Changes since v1 [1]: Incorporates feedback received prior to April 24.
1/ Ingo said [2]:
"So why on earth is this whole concept and the naming itself
('drivers/block/nd/' stands for 'NFIT Defined', apparently)
revolving around a specific 'firmware' mindset and revolving
around specific, weirdly named, overly complicated looking
firmware interfaces that come with their own new weird
glossary??"
Indeed, we of course consulted the NFIT specification to determine
the shape of the sub-system, but then let its terms and data
structures permeate too deep into the implementation. That is fixed
now with all NFIT specifics factored out into acpi.c. The NFIT is no
longer required reading to review libnd. Only three concepts are
needed:
i/ PMEM - contiguous memory range where cpu stores are
persistent once they are flushed through the memory
controller.
ii/ BLK - mmio apertures (sliding windows) that can be
programmed to access an aperture's-worth of persistent
media at a time.
iii/ DPA - "dimm-physical-address", address space local to a
dimm. A dimm may provide both PMEM-mode and BLK-mode
access to a range of DPA. libnd manages allocation of DPA
to either PMEM or BLK-namespaces to resolve this aliasing.
The v1..v2 diffstat below shows the migration of nfit-specifics to
acpi.c and the new state of libnd being nfit-free. "nd" now only
refers to "non-volatile devices". Note, reworked documentation will
return once the review has settled.
Documentation/blockdev/nd.txt | 867 ---------------------
MAINTAINERS | 34 +-
arch/ia64/kernel/efi.c | 5 +-
arch/x86/kernel/e820.c | 11 +-
arch/x86/kernel/pmem.c | 2 +-
drivers/block/Makefile | 2 +-
drivers/block/nd/Kconfig | 135 ++--
drivers/block/nd/Makefile | 32 +-
drivers/block/nd/acpi.c | 1506 +++++++++++++++++++++++++++++++------
drivers/block/nd/acpi_nfit.h | 321 ++++++++
drivers/block/nd/blk.c | 27 +-
drivers/block/nd/btt.c | 6 +-
drivers/block/nd/btt_devs.c | 8 +-
drivers/block/nd/bus.c | 337 +++++----
drivers/block/nd/core.c | 574 +-------------
drivers/block/nd/dimm.c | 11 -
drivers/block/nd/dimm_devs.c | 292 ++-----
drivers/block/nd/e820.c | 100 +++
drivers/block/nd/libnd.h | 122 +++
drivers/block/nd/namespace_devs.c | 10 +-
drivers/block/nd/nd-private.h | 107 +--
drivers/block/nd/nd.h | 91 +--
drivers/block/nd/nfit.h | 238 ------
drivers/block/nd/pmem.c | 56 +-
drivers/block/nd/region.c | 78 +-
drivers/block/nd/region_devs.c | 783 +++----------------
drivers/block/nd/test/iomap.c | 86 +--
drivers/block/nd/test/nfit.c | 1115 +++++++++++++++------------
drivers/block/nd/test/nfit_test.h | 15 +-
include/uapi/linux/ndctl.h | 130 ++--
30 files changed, 3166 insertions(+), 3935 deletions(-)
delete mode 100644 Documentation/blockdev/nd.txt
create mode 100644 drivers/block/nd/acpi_nfit.h
create mode 100644 drivers/block/nd/e820.c
create mode 100644 drivers/block/nd/libnd.h
delete mode 100644 drivers/block/nd/nfit.h
[1]: https://lists.01.org/pipermail/linux-nvdimm/2015-April/000484.html
[2]: https://lists.01.org/pipermail/linux-nvdimm/2015-April/000520.html
2/ Christoph asked the pmem ida conversion to be moved to its own patch
(done), and to consider leaving the current pmem.c in drivers/block/.
Instead, I converted the e820-type-12 enabling to be the first
non-ACPI-NFIT based consumer of libnd. The new nd_e820 driver simply
registers e820-type-12 ranges as libnd PMEM regions. Among other
things this conversion enables BTT for these ranges. The alternative
is to move drivers/block/nd/nd.h internals out to include/linux/
which I think is worse.
3/ Toshi reported that the NFIT parsing fails to handle the case of a
PMEM range with a single-dimm (non-aliasing) interleave description.
Support for this case was added and is tested by default by the
nfit_test.1 configuration.
4/ Toshi reported that we should not be treating a missing _STA property
as a "dimm disabled by firmware" case. (fixed).
5/ Christoph noted that ND_ARCH_HAS_IOREMAP_CACHE needs to be moved to
arch code. It is gone for now and we'll revisit when adding cached
mappings back to the PMEM driver.
6/ Toshi mentioned that the presence of two different nd_bus_probe()
functions was confusing. (cleaned up).
7/ Robert asked for s/btt_checksum/nd_btt_checksum/ (done).
8/ Linda asked for nfit_test to honor dynamic cma reservations via the
cma= command line (done). The cma requirements have also been
reduced to 128M as only the simulated DAX regions need CMA. The rest
can use vmalloc().
---
Available here:
git://git.kernel.org/pub/scm/linux/kernel/git/djbw/nvdimm nd-v2
---
Dan Williams (18):
e820, efi: add ACPI 6.0 persistent memory types
libnd, nd_acpi: initial libnd infrastructure and NFIT support
nd_acpi, nfit-test: manufactured NFITs for interface development
libnd: ndctl class device, and nd bus attributes
libnd, nd_acpi: dimm/memory-devices
libnd: ndctl.h, the nd ioctl abi
libnd, nd_dimm: dimm driver and base libnd device-driver infrastructure
libnd, nd_acpi: regions (block-data-window, persistent memory, volatile memory)
libnd: support for legacy (non-aliasing) nvdimms
pmem: use ida
libnd, nd_pmem: add libnd support to the pmem driver
libnd, nd_acpi: add interleave-set state-tracking infrastructure
libnd: namespace indices: read and validate
libnd: pmem label sets and namespace instantiation.
libnd: blk labels and namespace instantiation
libnd: write pmem label set
libnd: write blk label set
libnd: infrastructure for btt devices
Ross Zwisler (1):
libnd, nd_acpi, nd_blk: driver for BLK-mode access persistent memory
Vishal Verma (1):
nd_btt: atomic sector updates
Documentation/blockdev/btt.txt | 273 ++++++
arch/arm64/kernel/efi.c | 1
arch/ia64/kernel/efi.c | 4
arch/x86/boot/compressed/eboot.c | 4
arch/x86/include/uapi/asm/e820.h | 1
arch/x86/kernel/e820.c | 26 +
arch/x86/kernel/pmem.c | 2
arch/x86/platform/efi/efi.c | 3
drivers/block/Kconfig | 13
drivers/block/Makefile | 2
drivers/block/nd/Kconfig | 129 +++
drivers/block/nd/Makefile | 41 +
drivers/block/nd/acpi.c | 1505 +++++++++++++++++++++++++++++++++
drivers/block/nd/acpi_nfit.h | 321 +++++++
drivers/block/nd/blk.c | 264 ++++++
drivers/block/nd/btt.c | 1423 +++++++++++++++++++++++++++++++
drivers/block/nd/btt.h | 185 ++++
drivers/block/nd/btt_devs.c | 443 ++++++++++
drivers/block/nd/bus.c | 770 +++++++++++++++++
drivers/block/nd/core.c | 471 ++++++++++
drivers/block/nd/dimm.c | 115 +++
drivers/block/nd/dimm_devs.c | 507 +++++++++++
drivers/block/nd/e820.c | 100 ++
drivers/block/nd/label.c | 925 ++++++++++++++++++++
drivers/block/nd/label.h | 143 +++
drivers/block/nd/libnd.h | 122 +++
drivers/block/nd/namespace_devs.c | 1701 +++++++++++++++++++++++++++++++++++++
drivers/block/nd/nd-private.h | 114 ++
drivers/block/nd/nd.h | 261 ++++++
drivers/block/nd/pmem.c | 114 ++
drivers/block/nd/region.c | 159 +++
drivers/block/nd/region_devs.c | 637 ++++++++++++++
drivers/block/nd/test/Makefile | 5
drivers/block/nd/test/iomap.c | 151 +++
drivers/block/nd/test/nfit.c | 1131 +++++++++++++++++++++++++
drivers/block/nd/test/nfit_test.h | 26 +
include/linux/efi.h | 3
include/linux/nd.h | 98 ++
include/uapi/linux/Kbuild | 1
include/uapi/linux/ndctl.h | 199 ++++
40 files changed, 12345 insertions(+), 48 deletions(-)
create mode 100644 Documentation/blockdev/btt.txt
create mode 100644 drivers/block/nd/Kconfig
create mode 100644 drivers/block/nd/Makefile
create mode 100644 drivers/block/nd/acpi.c
create mode 100644 drivers/block/nd/acpi_nfit.h
create mode 100644 drivers/block/nd/blk.c
create mode 100644 drivers/block/nd/btt.c
create mode 100644 drivers/block/nd/btt.h
create mode 100644 drivers/block/nd/btt_devs.c
create mode 100644 drivers/block/nd/bus.c
create mode 100644 drivers/block/nd/core.c
create mode 100644 drivers/block/nd/dimm.c
create mode 100644 drivers/block/nd/dimm_devs.c
create mode 100644 drivers/block/nd/e820.c
create mode 100644 drivers/block/nd/label.c
create mode 100644 drivers/block/nd/label.h
create mode 100644 drivers/block/nd/libnd.h
create mode 100644 drivers/block/nd/namespace_devs.c
create mode 100644 drivers/block/nd/nd-private.h
create mode 100644 drivers/block/nd/nd.h
rename drivers/block/{pmem.c => nd/pmem.c} (68%)
create mode 100644 drivers/block/nd/region.c
create mode 100644 drivers/block/nd/region_devs.c
create mode 100644 drivers/block/nd/test/Makefile
create mode 100644 drivers/block/nd/test/iomap.c
create mode 100644 drivers/block/nd/test/nfit.c
create mode 100644 drivers/block/nd/test/nfit_test.h
create mode 100644 include/linux/nd.h
create mode 100644 include/uapi/linux/ndctl.h
ACPI 6.0 formalizes e820-type-7 and efi-type-14 as persistent memory.
Mark it "reserved" and allow it to be claimed by a persistent memory
device driver.
This definition is in addition to the Linux kernel's existing type-12
definition that was recently added in support of shipping platforms with
NVDIMM support that predate ACPI 6.0 (which now classifies type-12 as
OEM reserved). We may choose to exploit this wealth of definitions for
NVDIMMs to differentiate E820_PRAM (type-12) from E820_PMEM (type-7).
One potential differentiation is that PMEM is not backed by struct page
by default in contrast to PRAM. For now, they are effectively treated
as aliases by the mm.
Note, /proc/iomem can be consulted for differentiating legacy
"Persistent RAM" E820_PRAM vs standard "Persistent I/O Memory"
E820_PMEM.
Cc: Boaz Harrosh <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Borislav Petkov <[email protected]>
Cc: H. Peter Anvin <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Acked-by: Andy Lutomirski <[email protected]>
Reviewed-by: Ross Zwisler <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
arch/arm64/kernel/efi.c | 1 +
arch/ia64/kernel/efi.c | 4 ++++
arch/x86/boot/compressed/eboot.c | 4 ++++
arch/x86/include/uapi/asm/e820.h | 1 +
arch/x86/kernel/e820.c | 26 +++++++++++++++++++++++---
arch/x86/platform/efi/efi.c | 3 +++
include/linux/efi.h | 3 ++-
7 files changed, 38 insertions(+), 4 deletions(-)
diff --git a/arch/arm64/kernel/efi.c b/arch/arm64/kernel/efi.c
index ab21e0d58278..9d4aa18f2a82 100644
--- a/arch/arm64/kernel/efi.c
+++ b/arch/arm64/kernel/efi.c
@@ -158,6 +158,7 @@ static __init int is_reserve_region(efi_memory_desc_t *md)
case EFI_BOOT_SERVICES_CODE:
case EFI_BOOT_SERVICES_DATA:
case EFI_CONVENTIONAL_MEMORY:
+ case EFI_PERSISTENT_MEMORY:
return 0;
default:
break;
diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c
index c52d7540dc05..9028bc268cd7 100644
--- a/arch/ia64/kernel/efi.c
+++ b/arch/ia64/kernel/efi.c
@@ -1223,6 +1223,10 @@ efi_initialize_iomem_resources(struct resource *code_resource,
flags |= IORESOURCE_DISABLED;
break;
+ case EFI_PERSISTENT_MEMORY:
+ name = "persistent";
+ break;
+
case EFI_RESERVED_TYPE:
case EFI_RUNTIME_SERVICES_CODE:
case EFI_RUNTIME_SERVICES_DATA:
diff --git a/arch/x86/boot/compressed/eboot.c b/arch/x86/boot/compressed/eboot.c
index ef17683484e9..dde5bf7726f4 100644
--- a/arch/x86/boot/compressed/eboot.c
+++ b/arch/x86/boot/compressed/eboot.c
@@ -1222,6 +1222,10 @@ static efi_status_t setup_e820(struct boot_params *params,
e820_type = E820_NVS;
break;
+ case EFI_PERSISTENT_MEMORY:
+ e820_type = E820_PMEM;
+ break;
+
default:
continue;
}
diff --git a/arch/x86/include/uapi/asm/e820.h b/arch/x86/include/uapi/asm/e820.h
index 960a8a9dc4ab..0f457e6eab18 100644
--- a/arch/x86/include/uapi/asm/e820.h
+++ b/arch/x86/include/uapi/asm/e820.h
@@ -32,6 +32,7 @@
#define E820_ACPI 3
#define E820_NVS 4
#define E820_UNUSABLE 5
+#define E820_PMEM 7
/*
* This is a non-standardized way to represent ADR or NVDIMM regions that
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 11cc7d54ec3f..d38b53a7e9b2 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -149,6 +149,7 @@ static void __init e820_print_type(u32 type)
case E820_UNUSABLE:
printk(KERN_CONT "unusable");
break;
+ case E820_PMEM:
case E820_PRAM:
printk(KERN_CONT "persistent (type %u)", type);
break;
@@ -919,10 +920,31 @@ static inline const char *e820_type_to_string(int e820_type)
case E820_NVS: return "ACPI Non-volatile Storage";
case E820_UNUSABLE: return "Unusable memory";
case E820_PRAM: return "Persistent RAM";
+ case E820_PMEM: return "Persistent I/O Memory";
default: return "reserved";
}
}
+static bool do_mark_busy(u32 type, struct resource *res)
+{
+ /* this is the legacy bios/dos rom-shadow + mmio region */
+ if (res->start < (1ULL<<20))
+ return true;
+
+ /*
+ * Treat persistent memory like device memory, i.e. reserve it
+ * for exclusive use of a driver
+ */
+ switch (type) {
+ case E820_RESERVED:
+ case E820_PRAM:
+ case E820_PMEM:
+ return false;
+ default:
+ return true;
+ }
+}
+
/*
* Mark e820 reserved areas as busy for the resource manager.
*/
@@ -952,9 +974,7 @@ void __init e820_reserve_resources(void)
* pci device BAR resource and insert them later in
* pcibios_resource_survey()
*/
- if (((e820.map[i].type != E820_RESERVED) &&
- (e820.map[i].type != E820_PRAM)) ||
- res->start < (1ULL<<20)) {
+ if (do_mark_busy(e820.map[i].type, res)) {
res->flags |= IORESOURCE_BUSY;
insert_resource(&iomem_resource, res);
}
diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c
index dbc8627a5cdf..a116e236ac3f 100644
--- a/arch/x86/platform/efi/efi.c
+++ b/arch/x86/platform/efi/efi.c
@@ -145,6 +145,9 @@ static void __init do_add_efi_memmap(void)
case EFI_UNUSABLE_MEMORY:
e820_type = E820_UNUSABLE;
break;
+ case EFI_PERSISTENT_MEMORY:
+ e820_type = E820_PMEM;
+ break;
default:
/*
* EFI_RESERVED_TYPE EFI_RUNTIME_SERVICES_CODE
diff --git a/include/linux/efi.h b/include/linux/efi.h
index cf7e431cbc73..28868504aa17 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -85,7 +85,8 @@ typedef struct {
#define EFI_MEMORY_MAPPED_IO 11
#define EFI_MEMORY_MAPPED_IO_PORT_SPACE 12
#define EFI_PAL_CODE 13
-#define EFI_MAX_MEMORY_TYPE 14
+#define EFI_PERSISTENT_MEMORY 14
+#define EFI_MAX_MEMORY_TYPE 15
/* Attribute values: */
#define EFI_MEMORY_UC ((u64)0x0000000000000001ULL) /* uncached */
1/ Autodetect an NFIT table for the ACPI namespace device with _HID of
"ACPI0012"
2/ libnd bus registration
The NFIT provided by ACPI is one possible method by which platforms will
discover NVDIMM resources. However, the intent of the nd_bus_descriptor
abstraction is to abstract "provider" specific details, leaving libnd
to be independent of the specific NVDIMM resource discovery mechanism.
This flexibility is later exploited later to implement custom-defined nd
buses.
Cc: <[email protected]>
Cc: Robert Moore <[email protected]>
Cc: Rafael J. Wysocki <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/Kconfig | 2
drivers/block/Makefile | 1
drivers/block/nd/Kconfig | 40 +++
drivers/block/nd/Makefile | 6 +
drivers/block/nd/acpi.c | 475 +++++++++++++++++++++++++++++++++++++++++
drivers/block/nd/acpi_nfit.h | 254 ++++++++++++++++++++++
drivers/block/nd/core.c | 67 ++++++
drivers/block/nd/libnd.h | 33 +++
drivers/block/nd/nd-private.h | 23 ++
9 files changed, 901 insertions(+)
create mode 100644 drivers/block/nd/Kconfig
create mode 100644 drivers/block/nd/Makefile
create mode 100644 drivers/block/nd/acpi.c
create mode 100644 drivers/block/nd/acpi_nfit.h
create mode 100644 drivers/block/nd/core.c
create mode 100644 drivers/block/nd/libnd.h
create mode 100644 drivers/block/nd/nd-private.h
diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig
index eb1fed5bd516..dfe40e5ca9bd 100644
--- a/drivers/block/Kconfig
+++ b/drivers/block/Kconfig
@@ -321,6 +321,8 @@ config BLK_DEV_NVME
To compile this driver as a module, choose M here: the
module will be called nvme.
+source "drivers/block/nd/Kconfig"
+
config BLK_DEV_SKD
tristate "STEC S1120 Block Driver"
depends on PCI
diff --git a/drivers/block/Makefile b/drivers/block/Makefile
index 9cc6c18a1c7e..07a6acecf4d8 100644
--- a/drivers/block/Makefile
+++ b/drivers/block/Makefile
@@ -24,6 +24,7 @@ obj-$(CONFIG_CDROM_PKTCDVD) += pktcdvd.o
obj-$(CONFIG_MG_DISK) += mg_disk.o
obj-$(CONFIG_SUNVDC) += sunvdc.o
obj-$(CONFIG_BLK_DEV_NVME) += nvme.o
+obj-$(CONFIG_ND_DEVICES) += nd/
obj-$(CONFIG_BLK_DEV_SKD) += skd.o
obj-$(CONFIG_BLK_DEV_OSD) += osdblk.o
diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
new file mode 100644
index 000000000000..6d5d6b732f82
--- /dev/null
+++ b/drivers/block/nd/Kconfig
@@ -0,0 +1,40 @@
+menuconfig ND_DEVICES
+ bool "NVDIMM Support"
+ depends on PHYS_ADDR_T_64BIT
+ help
+ Generic support for non-volatile memory devices including
+ ACPI-6-NFIT defined resources. On platforms that define an
+ NFIT, or otherwise can discover NVDIMM resources, a libnd
+ bus is registered to advertise PMEM (persistent memory)
+ namespaces (/dev/pmemX) and BLK (sliding mmio window(s))
+ namespaces (/dev/ndX). A PMEM namespace refers to a memory
+ resource that may span multiple DIMMs and support DAX (see
+ CONFIG_DAX). A BLK namespace refers to an NVDIMM control
+ region which exposes an mmio register set for windowed
+ access mode to non-volatile memory.
+
+if ND_DEVICES
+
+config LIBND
+ tristate "LIBND: libnd device driver support"
+ help
+ Platform agnostic device model for a libnd bus. Publishes
+ resources for a PMEM (persistent-memory) driver and/or BLK
+ (sliding mmio window(s)) driver to attach. Exposes a device
+ topology under a "ndX" bus device, a "/dev/ndctlX" bus-ioctl
+ message passing interface, and a "/dev/nmemX" dimm-ioctl
+ message interface for each memory device registered on the
+ bus. instance. A userspace library "ndctl" provides an API
+ to enumerate/manage this subsystem.
+
+config ND_ACPI
+ tristate "ACPI: NFIT to libnd bus support"
+ select LIBND
+ depends on ACPI
+ help
+ Infrastructure to probe ACPI 6 compliant platforms for
+ NVDIMMs (NFIT) and register a libnd device tree. In
+ addition to storage devices this also enables libnd craft
+ ACPI._DSM messages for platform/dimm configuration.
+
+endif
diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
new file mode 100644
index 000000000000..944b5947c0cb
--- /dev/null
+++ b/drivers/block/nd/Makefile
@@ -0,0 +1,6 @@
+obj-$(CONFIG_LIBND) += libnd.o
+obj-$(CONFIG_ND_ACPI) += nd_acpi.o
+
+nd_acpi-y := acpi.o
+
+libnd-y := core.o
diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
new file mode 100644
index 000000000000..9f0b24390d1b
--- /dev/null
+++ b/drivers/block/nd/acpi.c
@@ -0,0 +1,475 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#include <linux/list_sort.h>
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/acpi.h>
+#include "acpi_nfit.h"
+#include "libnd.h"
+
+static bool warn_checksum;
+module_param(warn_checksum, bool, S_IRUGO|S_IWUSR);
+MODULE_PARM_DESC(warn_checksum, "Turn checksum errors into warnings");
+
+enum {
+ NFIT_ACPI_NOTIFY_TABLE = 0x80,
+};
+
+static int nd_acpi_ctl(struct nd_bus_descriptor *nd_desc,
+ struct nd_dimm *nd_dimm, unsigned int cmd, void *buf,
+ unsigned int buf_len)
+{
+ return -ENOTTY;
+}
+
+static const char *spa_type_name(u16 type)
+{
+ switch (type) {
+ case NFIT_SPA_VOLATILE: return "volatile";
+ case NFIT_SPA_PM: return "pmem";
+ case NFIT_SPA_DCR: return "dimm-control-region";
+ case NFIT_SPA_BDW: return "block-data-window";
+ default: return "unknown";
+ }
+}
+
+static int nfit_spa_type(struct acpi_nfit_spa *spa)
+{
+ if (memcmp(&nfit_spa_uuid_volatile, spa->type_uuid, 16) == 0)
+ return NFIT_SPA_VOLATILE;
+
+ if (memcmp(&nfit_spa_uuid_pm, spa->type_uuid, 16) == 0)
+ return NFIT_SPA_PM;
+
+ if (memcmp(&nfit_spa_uuid_dcr, spa->type_uuid, 16) == 0)
+ return NFIT_SPA_DCR;
+
+ if (memcmp(&nfit_spa_uuid_bdw, spa->type_uuid, 16) == 0)
+ return NFIT_SPA_BDW;
+
+ if (memcmp(&nfit_spa_uuid_vdisk, spa->type_uuid, 16) == 0)
+ return NFIT_SPA_VDISK;
+
+ if (memcmp(&nfit_spa_uuid_vcd, spa->type_uuid, 16) == 0)
+ return NFIT_SPA_VCD;
+
+ if (memcmp(&nfit_spa_uuid_pdisk, spa->type_uuid, 16) == 0)
+ return NFIT_SPA_PDISK;
+
+ if (memcmp(&nfit_spa_uuid_pcd, spa->type_uuid, 16) == 0)
+ return NFIT_SPA_PCD;
+
+ return -1;
+}
+
+struct nfit_table_header {
+ __le16 type;
+ __le16 length;
+};
+
+static void *add_table(struct acpi_nfit_desc *acpi_desc, void *table, const void *end)
+{
+ struct device *dev = acpi_desc->dev;
+ struct nfit_table_header *hdr;
+ void *err = ERR_PTR(-ENOMEM);
+
+ if (table >= end)
+ return NULL;
+
+ hdr = (struct nfit_table_header *) table;
+ switch (hdr->type) {
+ case NFIT_TABLE_SPA: {
+ struct nfit_spa *nfit_spa = devm_kzalloc(dev, sizeof(*nfit_spa),
+ GFP_KERNEL);
+ struct acpi_nfit_spa *spa = table;
+
+ if (!nfit_spa)
+ return err;
+ INIT_LIST_HEAD(&nfit_spa->list);
+ nfit_spa->spa = spa;
+ list_add_tail(&nfit_spa->list, &acpi_desc->spas);
+ dev_dbg(dev, "%s: spa index: %d type: %s\n", __func__,
+ spa->spa_index,
+ spa_type_name(nfit_spa_type(spa)));
+ break;
+ }
+ case NFIT_TABLE_MEM: {
+ struct nfit_memdev *nfit_memdev = devm_kzalloc(dev,
+ sizeof(*nfit_memdev), GFP_KERNEL);
+ struct acpi_nfit_memdev *memdev = table;
+
+ if (!nfit_memdev)
+ return err;
+ INIT_LIST_HEAD(&nfit_memdev->list);
+ nfit_memdev->memdev = memdev;
+ list_add_tail(&nfit_memdev->list, &acpi_desc->memdevs);
+ dev_dbg(dev, "%s: memdev handle: %#x spa: %d dcr: %d\n",
+ __func__, memdev->nfit_handle, memdev->spa_index,
+ memdev->dcr_index);
+ break;
+ }
+ case NFIT_TABLE_DCR: {
+ struct nfit_dcr *nfit_dcr = devm_kzalloc(dev, sizeof(*nfit_dcr),
+ GFP_KERNEL);
+ struct acpi_nfit_dcr *dcr = table;
+
+ if (!nfit_dcr)
+ return err;
+ INIT_LIST_HEAD(&nfit_dcr->list);
+ nfit_dcr->dcr = dcr;
+ list_add_tail(&nfit_dcr->list, &acpi_desc->dcrs);
+ dev_dbg(dev, "%s: dcr index: %d num_bcw: %d\n", __func__,
+ dcr->dcr_index, dcr->num_bcw);
+ break;
+ }
+ case NFIT_TABLE_BDW: {
+ struct nfit_bdw *nfit_bdw = devm_kzalloc(dev, sizeof(*nfit_bdw),
+ GFP_KERNEL);
+ struct acpi_nfit_bdw *bdw = table;
+
+ if (!nfit_bdw)
+ return err;
+ INIT_LIST_HEAD(&nfit_bdw->list);
+ nfit_bdw->bdw = bdw;
+ list_add_tail(&nfit_bdw->list, &acpi_desc->bdws);
+ dev_dbg(dev, "%s: bdw dcr: %d num_bdw: %d\n", __func__,
+ bdw->dcr_index, bdw->num_bdw);
+ break;
+ }
+ /* TODO */
+ case NFIT_TABLE_IDT:
+ dev_dbg(dev, "%s: idt\n", __func__);
+ break;
+ case NFIT_TABLE_FLUSH:
+ dev_dbg(dev, "%s: flush\n", __func__);
+ break;
+ case NFIT_TABLE_SMBIOS:
+ dev_dbg(dev, "%s: smbios\n", __func__);
+ break;
+ default:
+ dev_err(dev, "unknown table '%d' parsing nfit\n", hdr->type);
+ return ERR_PTR(-ENXIO);
+ }
+
+ return table + hdr->length;
+}
+
+static void nfit_mem_find_spa_bdw(struct acpi_nfit_desc *acpi_desc,
+ struct nfit_mem *nfit_mem)
+{
+ u32 nfit_handle = __to_nfit_memdev(nfit_mem)->nfit_handle;
+ u16 dcr_index = nfit_mem->dcr->dcr_index;
+ struct nfit_spa *nfit_spa;
+
+ list_for_each_entry(nfit_spa, &acpi_desc->spas, list) {
+ u16 spa_index = nfit_spa->spa->spa_index;
+ int type = nfit_spa_type(nfit_spa->spa);
+ struct nfit_memdev *nfit_memdev;
+
+ if (type != NFIT_SPA_BDW)
+ continue;
+
+ list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) {
+ if (nfit_memdev->memdev->spa_index != spa_index)
+ continue;
+ if (nfit_memdev->memdev->nfit_handle != nfit_handle)
+ continue;
+ if (nfit_memdev->memdev->dcr_index != dcr_index)
+ continue;
+
+ nfit_mem->spa_bdw = nfit_spa->spa;
+ return;
+ }
+ }
+
+ dev_dbg(acpi_desc->dev, "SPA-BDW not found for SPA-DCR %d\n",
+ nfit_mem->spa_dcr->spa_index);
+ nfit_mem->bdw = NULL;
+}
+
+static int nfit_mem_add(struct acpi_nfit_desc *acpi_desc,
+ struct nfit_mem *nfit_mem, struct acpi_nfit_spa *spa)
+{
+ u16 dcr_index = __to_nfit_memdev(nfit_mem)->dcr_index;
+ struct nfit_dcr *nfit_dcr;
+ struct nfit_bdw *nfit_bdw;
+
+ list_for_each_entry(nfit_dcr, &acpi_desc->dcrs, list) {
+ if (nfit_dcr->dcr->dcr_index != dcr_index)
+ continue;
+ nfit_mem->dcr = nfit_dcr->dcr;
+ break;
+ }
+
+ if (!nfit_mem->dcr) {
+ dev_dbg(acpi_desc->dev, "SPA %d missing:%s%s\n", spa->spa_index,
+ __to_nfit_memdev(nfit_mem) ? "" : " MEMDEV",
+ nfit_mem->dcr ? "" : " DCR");
+ return -ENODEV;
+ }
+
+ /*
+ * We've found enough to create an nd_dimm, optionally
+ * find an associated BDW
+ */
+ list_add(&nfit_mem->list, &acpi_desc->dimms);
+
+ list_for_each_entry(nfit_bdw, &acpi_desc->bdws, list) {
+ if (nfit_bdw->bdw->dcr_index != dcr_index)
+ continue;
+ nfit_mem->bdw = nfit_bdw->bdw;
+ break;
+ }
+
+ if (!nfit_mem->bdw)
+ return 0;
+
+ nfit_mem_find_spa_bdw(acpi_desc, nfit_mem);
+ return 0;
+}
+
+static int nfit_mem_dcr_init(struct acpi_nfit_desc *acpi_desc,
+ struct acpi_nfit_spa *spa)
+{
+ struct nfit_mem *nfit_mem, *found;
+ struct nfit_memdev *nfit_memdev;
+ int type = nfit_spa_type(spa);
+ u16 dcr_index;
+
+ switch (type) {
+ case NFIT_SPA_DCR:
+ case NFIT_SPA_PM:
+ break;
+ default:
+ return 0;
+ }
+
+ list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) {
+ int rc;
+
+ if (nfit_memdev->memdev->spa_index != spa->spa_index)
+ continue;
+ found = NULL;
+ dcr_index = nfit_memdev->memdev->dcr_index;
+ list_for_each_entry(nfit_mem, &acpi_desc->dimms, list)
+ if (__to_nfit_memdev(nfit_mem)->dcr_index == dcr_index) {
+ found = nfit_mem;
+ break;
+ }
+
+ if (found)
+ nfit_mem = found;
+ else {
+ nfit_mem = devm_kzalloc(acpi_desc->dev,
+ sizeof(*nfit_mem), GFP_KERNEL);
+ if (!nfit_mem)
+ return -ENOMEM;
+ INIT_LIST_HEAD(&nfit_mem->list);
+ }
+
+ if (type == NFIT_SPA_DCR) {
+ /* multiple dimms may share a SPA when interleaved */
+ nfit_mem->spa_dcr = spa;
+ nfit_mem->memdev_dcr = nfit_memdev->memdev;
+ } else {
+ /*
+ * A single dimm may belong to multiple SPA-PM
+ * ranges, record at least one in addition to
+ * any SPA-DCR range.
+ */
+ nfit_mem->memdev_pmem = nfit_memdev->memdev;
+ }
+
+ if (found)
+ continue;
+
+ rc = nfit_mem_add(acpi_desc, nfit_mem, spa);
+ if (rc)
+ return rc;
+ }
+
+ return 0;
+}
+
+static int nfit_mem_cmp(void *priv, struct list_head *__a, struct list_head *__b)
+{
+ struct nfit_mem *a = container_of(__a, typeof(*a), list);
+ struct nfit_mem *b = container_of(__b, typeof(*b), list);
+ u32 handleA, handleB;
+
+ handleA = __to_nfit_memdev(a)->nfit_handle;
+ handleB = __to_nfit_memdev(b)->nfit_handle;
+ if (handleA < handleB)
+ return -1;
+ else if (handleA > handleB)
+ return 1;
+ return 0;
+}
+
+static int nfit_mem_init(struct acpi_nfit_desc *acpi_desc)
+{
+ struct nfit_spa *nfit_spa;
+
+ /*
+ * For each SPA-DCR or SPA-PMEM address range find its
+ * corresponding MEMDEV(s). From each MEMDEV find the
+ * corresponding DCR. Then, if we're operating on a SPA-DCR,
+ * try to find a SPA-BDW and a corresponding BDW that references
+ * the DCR. Throw it all into an nfit_mem object. Note, that
+ * BDWs are optional.
+ */
+ list_for_each_entry(nfit_spa, &acpi_desc->spas, list) {
+ int rc;
+
+ rc = nfit_mem_dcr_init(acpi_desc, nfit_spa->spa);
+ if (rc)
+ return rc;
+ }
+
+ list_sort(NULL, &acpi_desc->dimms, nfit_mem_cmp);
+
+ return 0;
+}
+
+static int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
+{
+ struct device *dev = acpi_desc->dev;
+ const void *end;
+ u8 *data, sum;
+ acpi_size i;
+
+ INIT_LIST_HEAD(&acpi_desc->spas);
+ INIT_LIST_HEAD(&acpi_desc->dcrs);
+ INIT_LIST_HEAD(&acpi_desc->bdws);
+ INIT_LIST_HEAD(&acpi_desc->memdevs);
+ INIT_LIST_HEAD(&acpi_desc->dimms);
+
+ data = (u8 *) acpi_desc->nfit;
+ for (i = 0, sum = 0; i < sz; i++)
+ sum += readb(data + i);
+ if (sum != 0 && !warn_checksum) {
+ dev_dbg(dev, "%s: nfit checksum failure\n", __func__);
+ return -ENXIO;
+ }
+ WARN_TAINT_ONCE(sum != 0, TAINT_FIRMWARE_WORKAROUND,
+ "nfit checksum failure, continuing...\n");
+
+ end = data + sz;
+ data += sizeof(struct acpi_nfit);
+ while (!IS_ERR_OR_NULL(data))
+ data = add_table(acpi_desc, data, end);
+
+ if (IS_ERR(data)) {
+ dev_dbg(dev, "%s: nfit table parsing error: %ld\n", __func__,
+ PTR_ERR(data));
+ return PTR_ERR(data);
+ }
+
+ if (nfit_mem_init(acpi_desc) != 0)
+ return -ENOMEM;
+
+ return 0;
+}
+
+static int nd_acpi_add(struct acpi_device *adev)
+{
+ struct nd_bus_descriptor *nd_desc;
+ struct acpi_nfit_desc *acpi_desc;
+ struct device *dev = &adev->dev;
+ struct acpi_table_header *tbl;
+ acpi_status status = AE_OK;
+ acpi_size sz;
+ int rc;
+
+ status = acpi_get_table_with_size("NFIT", 0, &tbl, &sz);
+ if (ACPI_FAILURE(status)) {
+ dev_err(dev, "failed to find NFIT\n");
+ return -ENXIO;
+ }
+
+ acpi_desc = devm_kzalloc(dev, sizeof(*acpi_desc), GFP_KERNEL);
+ if (!acpi_desc)
+ return -ENOMEM;
+
+ dev_set_drvdata(dev, acpi_desc);
+ acpi_desc->dev = dev;
+ acpi_desc->nfit = (struct acpi_nfit *) tbl;
+ nd_desc = &acpi_desc->nd_desc;
+ nd_desc->provider_name = "ACPI.NFIT";
+ nd_desc->ndctl = nd_acpi_ctl;
+
+ acpi_desc->nd_bus = nd_bus_register(dev, nd_desc);
+ if (!acpi_desc->nd_bus)
+ return -ENXIO;
+
+ rc = nd_acpi_nfit_init(acpi_desc, sz);
+ if (rc) {
+ nd_bus_unregister(acpi_desc->nd_bus);
+ return rc;
+ }
+ return 0;
+}
+
+static int nd_acpi_remove(struct acpi_device *adev)
+{
+ struct acpi_nfit_desc *acpi_desc = dev_get_drvdata(&adev->dev);
+
+ nd_bus_unregister(acpi_desc->nd_bus);
+ return 0;
+}
+
+static void nd_acpi_notify(struct acpi_device *adev, u32 event)
+{
+ /* TODO: handle ACPI_NOTIFY_BUS_CHECK notification */
+ dev_dbg(&adev->dev, "%s: event: %d\n", __func__, event);
+}
+
+static const struct acpi_device_id nd_acpi_ids[] = {
+ { "ACPI0012", 0 },
+ { "", 0 },
+};
+MODULE_DEVICE_TABLE(acpi, nd_acpi_ids);
+
+static struct acpi_driver nd_acpi_driver = {
+ .name = KBUILD_MODNAME,
+ .ids = nd_acpi_ids,
+ .flags = ACPI_DRIVER_ALL_NOTIFY_EVENTS,
+ .ops = {
+ .add = nd_acpi_add,
+ .remove = nd_acpi_remove,
+ .notify = nd_acpi_notify
+ },
+};
+
+static __init int nd_acpi_init(void)
+{
+ BUILD_BUG_ON(sizeof(struct acpi_nfit) != 40);
+ BUILD_BUG_ON(sizeof(struct acpi_nfit_spa) != 56);
+ BUILD_BUG_ON(sizeof(struct acpi_nfit_memdev) != 48);
+ BUILD_BUG_ON(sizeof(struct acpi_nfit_idt) != 16);
+ BUILD_BUG_ON(sizeof(struct acpi_nfit_smbios) != 8);
+ BUILD_BUG_ON(sizeof(struct acpi_nfit_dcr) != 80);
+ BUILD_BUG_ON(sizeof(struct acpi_nfit_bdw) != 40);
+
+ return acpi_bus_register_driver(&nd_acpi_driver);
+}
+
+static __exit void nd_acpi_exit(void)
+{
+ acpi_bus_unregister_driver(&nd_acpi_driver);
+}
+
+module_init(nd_acpi_init);
+module_exit(nd_acpi_exit);
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Intel Corporation");
diff --git a/drivers/block/nd/acpi_nfit.h b/drivers/block/nd/acpi_nfit.h
new file mode 100644
index 000000000000..e0b0f12736bf
--- /dev/null
+++ b/drivers/block/nd/acpi_nfit.h
@@ -0,0 +1,254 @@
+/*
+ * NVDIMM Firmware Interface Table - NFIT
+ *
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#ifndef __NFIT_H__
+#define __NFIT_H__
+#include <linux/types.h>
+#include <linux/uuid.h>
+#include <linux/acpi.h>
+#include "libnd.h"
+
+static const uuid_le nfit_spa_uuid_volatile __maybe_unused = UUID_LE(0x7305944f,
+ 0xfdda, 0x44e3, 0xb1, 0x6c, 0x3f, 0x22, 0xd2, 0x52, 0xe5, 0xd0);
+
+static const uuid_le nfit_spa_uuid_pm __maybe_unused = UUID_LE(0x66f0d379,
+ 0xb4f3, 0x4074, 0xac, 0x43, 0x0d, 0x33, 0x18, 0xb7, 0x8c, 0xdb);
+
+static const uuid_le nfit_spa_uuid_dcr __maybe_unused = UUID_LE(0x92f701f6,
+ 0x13b4, 0x405d, 0x91, 0x0b, 0x29, 0x93, 0x67, 0xe8, 0x23, 0x4c);
+
+static const uuid_le nfit_spa_uuid_bdw __maybe_unused = UUID_LE(0x91af0530,
+ 0x5d86, 0x470e, 0xa6, 0xb0, 0x0a, 0x2d, 0xb9, 0x40, 0x82, 0x49);
+
+static const uuid_le nfit_spa_uuid_vdisk __maybe_unused = UUID_LE(0x77ab535a,
+ 0x45fc, 0x624b, 0x55, 0x60, 0xf7, 0xb2, 0x81, 0xd1, 0xf9, 0x6e);
+
+static const uuid_le nfit_spa_uuid_vcd __maybe_unused = UUID_LE(0x3d5abd30,
+ 0x4175, 0x87ce, 0x6d, 0x64, 0xd2, 0xad, 0xe5, 0x23, 0xc4, 0xbb);
+
+static const uuid_le nfit_spa_uuid_pdisk __maybe_unused = UUID_LE(0x5cea02c9,
+ 0x4d07, 0x69d3, 0x26, 0x9f, 0x44, 0x96, 0xfb, 0xe0, 0x96, 0xf9);
+
+static const uuid_le nfit_spa_uuid_pcd __maybe_unused = UUID_LE(0x08018188,
+ 0x42cd, 0xbb48, 0x10, 0x0f, 0x53, 0x87, 0xd5, 0x3d, 0xed, 0x3d);
+
+enum {
+ NFIT_TABLE_SPA = 0,
+ NFIT_TABLE_MEM = 1,
+ NFIT_TABLE_IDT = 2,
+ NFIT_TABLE_SMBIOS = 3,
+ NFIT_TABLE_DCR = 4,
+ NFIT_TABLE_BDW = 5,
+ NFIT_TABLE_FLUSH = 6,
+ NFIT_SPA_VOLATILE = 0,
+ NFIT_SPA_PM = 1,
+ NFIT_SPA_DCR = 2,
+ NFIT_SPA_BDW = 3,
+ NFIT_SPA_VDISK = 4,
+ NFIT_SPA_VCD = 5,
+ NFIT_SPA_PDISK = 6,
+ NFIT_SPA_PCD = 7,
+ NFIT_SPAF_DCR_HOT_ADD = 1 << 0,
+ NFIT_SPAF_PDVALID = 1 << 1,
+ NFIT_MEMF_SAVE_FAIL = 1 << 0,
+ NFIT_MEMF_RESTORE_FAIL = 1 << 1,
+ NFIT_MEMF_FLUSH_FAIL = 1 << 2,
+ NFIT_MEMF_UNARMED = 1 << 3,
+ NFIT_MEMF_NOTIFY_SMART = 1 << 4,
+ NFIT_MEMF_SMART_READY = 1 << 5,
+ NFIT_DCRF_BUFFERED = 1 << 0,
+};
+
+/**
+ * struct acpi_nfit - Nvdimm Firmware Interface Table
+ * @signature: "NFIT"
+ * @length: sum of size of this table plus all appended subtables
+ */
+struct acpi_nfit {
+ u8 signature[4];
+ u32 length;
+ u8 revision;
+ u8 checksum;
+ u8 oemid[6];
+ u64 oem_tbl_id;
+ u32 oem_revision;
+ u32 creator_id;
+ u32 creator_revision;
+ u32 reserved;
+};
+
+/**
+ * struct acpi_nfit_spa - System Physical Address Range Descriptor Table
+ */
+struct acpi_nfit_spa {
+ u16 type;
+ u16 length;
+ u16 spa_index;
+ u16 flags;
+ u32 reserved;
+ u32 proximity_domain;
+ u8 type_uuid[16];
+ u64 spa_base;
+ u64 spa_length;
+ u64 mem_attr;
+};
+
+/**
+ * struct acpi_nfit_mem - Memory Device to SPA Mapping Table
+ */
+struct acpi_nfit_memdev {
+ u16 type;
+ u16 length;
+ u32 nfit_handle;
+ u16 phys_id;
+ u16 region_id;
+ u16 spa_index;
+ u16 dcr_index;
+ u64 region_len;
+ u64 region_spa_offset;
+ u64 region_dpa;
+ u16 idt_index;
+ u16 interleave_ways;
+ u16 flags;
+ u16 reserved;
+};
+
+/**
+ * struct acpi_nfit_idt - Interleave description Table
+ */
+struct acpi_nfit_idt {
+ u16 type;
+ u16 length;
+ u16 idt_index;
+ u16 reserved;
+ u32 num_lines;
+ u32 line_size;
+ u32 line_offset[0];
+};
+
+/**
+ * struct acpi_nfit_smbios - SMBIOS Management Information Table
+ */
+struct acpi_nfit_smbios {
+ u16 type;
+ u16 length;
+ u32 reserved;
+ u8 data[0];
+};
+
+/**
+ * struct acpi_nfit_dcr - NVDIMM Control Region Table
+ * @fic: Format Interface Code
+ * @cmd_offset: command registers relative to block control window
+ * @status_offset: status registers relative to block control window
+ */
+struct acpi_nfit_dcr {
+ u16 type;
+ u16 length;
+ u16 dcr_index;
+ u16 vendor_id;
+ u16 device_id;
+ u16 revision_id;
+ u16 sub_vendor_id;
+ u16 sub_device_id;
+ u16 sub_revision_id;
+ u8 reserved[6];
+ u32 serial_number;
+ u16 fic;
+ u16 num_bcw;
+ u64 bcw_size;
+ u64 cmd_offset;
+ u64 cmd_size;
+ u64 status_offset;
+ u64 status_size;
+ u16 flags;
+ u8 reserved2[6];
+};
+
+/**
+ * struct acpi_nfit_bdw - NVDIMM Block Data Window Region Table
+ */
+struct acpi_nfit_bdw {
+ u16 type;
+ u16 length;
+ u16 dcr_index;
+ u16 num_bdw;
+ u64 bdw_offset;
+ u64 bdw_size;
+ u64 blk_capacity;
+ u64 blk_offset;
+};
+
+/**
+ * struct acpi_nfit_flush - Flush Hint Address Structure
+ */
+struct acpi_nfit_flush {
+ u16 type;
+ u16 length;
+ u32 nfit_handle;
+ u16 num_hints;
+ u8 reserved[6];
+ u64 hint_addr[0];
+};
+
+struct nfit_spa {
+ struct acpi_nfit_spa *spa;
+ struct list_head list;
+};
+
+struct nfit_dcr {
+ struct acpi_nfit_dcr *dcr;
+ struct list_head list;
+};
+
+struct nfit_bdw {
+ struct acpi_nfit_bdw *bdw;
+ struct list_head list;
+};
+
+struct nfit_memdev {
+ struct acpi_nfit_memdev *memdev;
+ struct list_head list;
+};
+
+/* assembled tables for a given dimm/memory-device */
+struct nfit_mem {
+ struct acpi_nfit_memdev *memdev_dcr;
+ struct acpi_nfit_memdev *memdev_pmem;
+ struct acpi_nfit_dcr *dcr;
+ struct acpi_nfit_bdw *bdw;
+ struct acpi_nfit_spa *spa_dcr;
+ struct acpi_nfit_spa *spa_bdw;
+ struct list_head list;
+};
+
+struct acpi_nfit_desc {
+ struct nd_bus_descriptor nd_desc;
+ struct acpi_nfit *nfit;
+ struct list_head memdevs;
+ struct list_head dimms;
+ struct list_head spas;
+ struct list_head dcrs;
+ struct list_head bdws;
+ struct nd_bus *nd_bus;
+ struct device *dev;
+};
+
+static inline struct acpi_nfit_memdev *__to_nfit_memdev(struct nfit_mem *nfit_mem)
+{
+ if (nfit_mem->memdev_dcr)
+ return nfit_mem->memdev_dcr;
+ return nfit_mem->memdev_pmem;
+}
+#endif /* __NFIT_H__ */
diff --git a/drivers/block/nd/core.c b/drivers/block/nd/core.c
new file mode 100644
index 000000000000..3cccdbc0f3b7
--- /dev/null
+++ b/drivers/block/nd/core.c
@@ -0,0 +1,67 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#include <linux/export.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/slab.h>
+#include "nd-private.h"
+#include "libnd.h"
+
+static DEFINE_IDA(nd_ida);
+
+static void nd_bus_release(struct device *dev)
+{
+ struct nd_bus *nd_bus = container_of(dev, struct nd_bus, dev);
+
+ ida_simple_remove(&nd_ida, nd_bus->id);
+ kfree(nd_bus);
+}
+
+struct nd_bus *nd_bus_register(struct device *parent,
+ struct nd_bus_descriptor *nd_desc)
+{
+ struct nd_bus *nd_bus = kzalloc(sizeof(*nd_bus), GFP_KERNEL);
+ int rc;
+
+ if (!nd_bus)
+ return NULL;
+ nd_bus->id = ida_simple_get(&nd_ida, 0, 0, GFP_KERNEL);
+ if (nd_bus->id < 0) {
+ kfree(nd_bus);
+ return NULL;
+ }
+ nd_bus->nd_desc = nd_desc;
+ nd_bus->dev.parent = parent;
+ nd_bus->dev.release = nd_bus_release;
+ dev_set_name(&nd_bus->dev, "ndbus%d", nd_bus->id);
+ rc = device_register(&nd_bus->dev);
+ if (rc) {
+ dev_dbg(&nd_bus->dev, "device registration failed: %d\n", rc);
+ put_device(&nd_bus->dev);
+ return NULL;
+ }
+
+ return nd_bus;
+}
+EXPORT_SYMBOL_GPL(nd_bus_register);
+
+void nd_bus_unregister(struct nd_bus *nd_bus)
+{
+ if (!nd_bus)
+ return;
+ device_unregister(&nd_bus->dev);
+}
+EXPORT_SYMBOL_GPL(nd_bus_unregister);
+
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Intel Corporation");
diff --git a/drivers/block/nd/libnd.h b/drivers/block/nd/libnd.h
new file mode 100644
index 000000000000..163832937e9c
--- /dev/null
+++ b/drivers/block/nd/libnd.h
@@ -0,0 +1,33 @@
+/*
+ * libnd - Non-volatile-memory Devices Subsystem
+ *
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#ifndef __LIBND_H__
+#define __LIBND_H__
+struct nd_dimm;
+struct nd_bus_descriptor;
+typedef int (*ndctl_fn)(struct nd_bus_descriptor *nd_desc,
+ struct nd_dimm *nd_dimm, unsigned int cmd, void *buf,
+ unsigned int buf_len);
+
+struct nd_bus_descriptor {
+ unsigned long dsm_mask;
+ char *provider_name;
+ ndctl_fn ndctl;
+};
+
+struct nd_bus;
+struct nd_bus *nd_bus_register(struct device *parent,
+ struct nd_bus_descriptor *nfit_desc);
+void nd_bus_unregister(struct nd_bus *nd_bus);
+#endif /* __LIBND_H__ */
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
new file mode 100644
index 000000000000..3dbab29fa0f9
--- /dev/null
+++ b/drivers/block/nd/nd-private.h
@@ -0,0 +1,23 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#ifndef __ND_PRIVATE_H__
+#define __ND_PRIVATE_H__
+#include <linux/device.h>
+#include "libnd.h"
+
+struct nd_bus {
+ struct nd_bus_descriptor *nd_desc;
+ struct device dev;
+ int id;
+};
+#endif /* __ND_PRIVATE_H__ */
Manually create and register NFITs to describe 2 topologies. Topology1
is an advanced plausible configuration for BLK/PMEM aliased NVDIMMs.
Topology2 is an example configuration for current platforms that only
ship with a persistent address range.
Kernel provider "nfit_test.0" produces an NFIT with the following attributes:
(a) (b) DIMM BLK-REGION
+-------------------+--------+--------+--------+
+------+ | pm0.0 | blk2.0 | pm1.0 | blk2.1 | 0 region2
| imc0 +--+- - - region0- - - +--------+ +--------+
+--+---+ | pm0.0 | blk3.0 | pm1.0 | blk3.1 | 1 region3
| +-------------------+--------v v--------+
+--+---+ | |
| cpu0 | region1
+--+---+ | |
| +----------------------------^ ^--------+
+--+---+ | blk4.0 | pm1.0 | blk4.0 | 2 region4
| imc1 +--+----------------------------| +--------+
+------+ | blk5.0 | pm1.0 | blk5.0 | 3 region5
+----------------------------+--------+--------+
*) In this layout we have four dimms and two memory controllers in one
socket. Each unique interface ("block" or "pmem") to DPA space
is identified by a region device with a dynamically assigned id.
*) The first portion of dimm0 and dimm1 are interleaved as REGION0.
A single "pmem" namespace is created in the REGION0-"spa"-range
that spans dimm0 and dimm1 with a user-specified name of "pm0.0".
Some of that interleaved "spa" range is reclaimed as "bdw"
accessed space starting at offset (a) into each dimm. In that
reclaimed space we create two "bdw" "namespaces" from REGION2 and
REGION3 where "blk2.0" and "blk3.0" are just human readable names
that could be set to any user-desired name in the label.
*) In the last portion of dimm0 and dimm1 we have an interleaved
"spa" range, REGION1, that spans those two dimms as well as dimm2
and dimm3. Some of REGION1 allocated to a "pmem" namespace named
"pm1.0" the rest is reclaimed in 4 "bdw" namespaces (for each
dimm in the interleave set), "blk2.1", "blk3.1", "blk4.0", and
"blk5.0".
*) The portion of dimm2 and dimm3 that do not participate in the
REGION1 interleaved "spa" range (i.e. the DPA address below
offset (b) are also included in the "blk4.0" and "blk5.0"
namespaces. Note, that this example shows that "bdw" namespaces
don't need to be contiguous in DPA-space.
Kernel provider "nfit_test.1" produces an NFIT with the following attributes:
region2
+---------------------+
|---------------------|
|| pm2.0 ||
|---------------------|
+---------------------+
*) Describes a simple system-physical-address range with no backing
dimm or interleave description.
Cc: <[email protected]>
Cc: Robert Moore <[email protected]>
Cc: Rafael J. Wysocki <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/Kconfig | 19 +
drivers/block/nd/Makefile | 15 +
drivers/block/nd/acpi.c | 3
drivers/block/nd/acpi_nfit.h | 11
drivers/block/nd/test/Makefile | 5
drivers/block/nd/test/iomap.c | 151 +++++
drivers/block/nd/test/nfit.c | 1025 +++++++++++++++++++++++++++++++++++++
drivers/block/nd/test/nfit_test.h | 26 +
8 files changed, 1254 insertions(+), 1 deletion(-)
create mode 100644 drivers/block/nd/test/Makefile
create mode 100644 drivers/block/nd/test/iomap.c
create mode 100644 drivers/block/nd/test/nfit.c
create mode 100644 drivers/block/nd/test/nfit_test.h
diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
index 6d5d6b732f82..09f0135147ca 100644
--- a/drivers/block/nd/Kconfig
+++ b/drivers/block/nd/Kconfig
@@ -37,4 +37,23 @@ config ND_ACPI
addition to storage devices this also enables libnd craft
ACPI._DSM messages for platform/dimm configuration.
+config NFIT_TEST
+ tristate "NFIT TEST: Manufactured NFIT for interface testing"
+ depends on DMA_CMA
+ depends on LIBND=m
+ depends on ND_ACPI
+ depends on m
+ help
+ For development purposes register a manufactured
+ NFIT table to verify the resulting device model topology.
+ Note, this module arranges for ioremap_cache() to be
+ overridden locally to allow simulation of system-memory as an
+ io-memory-resource.
+
+ Note, this test expects to be able to find at least 256MB of
+ CMA space (CONFIG_CMA_SIZE_MBYTES, cma=) or it will fail to
+ load.
+
+ Say N unless you are doing development of the 'nd' subsystem.
+
endif
diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
index 944b5947c0cb..cf064db92589 100644
--- a/drivers/block/nd/Makefile
+++ b/drivers/block/nd/Makefile
@@ -1,5 +1,20 @@
+ifdef CONFIG_NFIT_TEST
+# This obviously will cause symbol collisions if another
+# driver/sub-system attempts a similar mocked io-memory implementation.
+# When that happens we can either add a 'choice' kconfig option to
+# select one mocked instance at a time, or push for the linker to
+# include an option of the form "--wrap-prefix=<prefix>" to allow for
+# separate namespaces of mocked functions.
+ldflags-y += --wrap=ioremap_cache
+ldflags-y += --wrap=ioremap_nocache
+ldflags-y += --wrap=iounmap
+ldflags-y += --wrap=__request_region
+ldflags-y += --wrap=__release_region
+endif
+
obj-$(CONFIG_LIBND) += libnd.o
obj-$(CONFIG_ND_ACPI) += nd_acpi.o
+obj-$(CONFIG_NFIT_TEST) += test/
nd_acpi-y := acpi.o
diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
index 9f0b24390d1b..54344ef9c837 100644
--- a/drivers/block/nd/acpi.c
+++ b/drivers/block/nd/acpi.c
@@ -341,7 +341,7 @@ static int nfit_mem_init(struct acpi_nfit_desc *acpi_desc)
return 0;
}
-static int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
+int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
{
struct device *dev = acpi_desc->dev;
const void *end;
@@ -380,6 +380,7 @@ static int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
return 0;
}
+EXPORT_SYMBOL_GPL(nd_acpi_nfit_init);
static int nd_acpi_add(struct acpi_device *adev)
{
diff --git a/drivers/block/nd/acpi_nfit.h b/drivers/block/nd/acpi_nfit.h
index e0b0f12736bf..a26f69e32244 100644
--- a/drivers/block/nd/acpi_nfit.h
+++ b/drivers/block/nd/acpi_nfit.h
@@ -124,6 +124,15 @@ struct acpi_nfit_memdev {
u16 reserved;
};
+#define NFIT_DIMM_HANDLE(node, socket, imc, chan, dimm) \
+ (((node & 0xfff) << 16) | ((socket & 0xf) << 12) \
+ | ((imc & 0xf) << 8) | ((chan & 0xf) << 4) | (dimm & 0xf))
+#define NFIT_DIMM_NODE(handle) ((handle) >> 16 & 0xfff)
+#define NFIT_DIMM_SOCKET(handle) ((handle) >> 12 & 0xf)
+#define NFIT_DIMM_CHAN(handle) ((handle) >> 8 & 0xf)
+#define NFIT_DIMM_IMC(handle) ((handle) >> 4 & 0xf)
+#define NFIT_DIMM_DIMM(handle) ((handle) & 0xf)
+
/**
* struct acpi_nfit_idt - Interleave description Table
*/
@@ -251,4 +260,6 @@ static inline struct acpi_nfit_memdev *__to_nfit_memdev(struct nfit_mem *nfit_me
return nfit_mem->memdev_dcr;
return nfit_mem->memdev_pmem;
}
+
+int nd_acpi_nfit_init(struct acpi_nfit_desc *nfit, acpi_size sz);
#endif /* __NFIT_H__ */
diff --git a/drivers/block/nd/test/Makefile b/drivers/block/nd/test/Makefile
new file mode 100644
index 000000000000..c7f319cbd082
--- /dev/null
+++ b/drivers/block/nd/test/Makefile
@@ -0,0 +1,5 @@
+obj-$(CONFIG_NFIT_TEST) += nfit_test.o
+obj-$(CONFIG_NFIT_TEST) += nfit_test_iomap.o
+
+nfit_test-y := nfit.o
+nfit_test_iomap-y := iomap.o
diff --git a/drivers/block/nd/test/iomap.c b/drivers/block/nd/test/iomap.c
new file mode 100644
index 000000000000..c85a6f6ba559
--- /dev/null
+++ b/drivers/block/nd/test/iomap.c
@@ -0,0 +1,151 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#include <linux/rculist.h>
+#include <linux/export.h>
+#include <linux/ioport.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/io.h>
+#include "nfit_test.h"
+
+static LIST_HEAD(iomap_head);
+
+static struct iomap_ops {
+ nfit_test_lookup_fn nfit_test_lookup;
+ struct list_head list;
+} iomap_ops = {
+ .list = LIST_HEAD_INIT(iomap_ops.list),
+};
+
+void nfit_test_setup(nfit_test_lookup_fn lookup)
+{
+ iomap_ops.nfit_test_lookup = lookup;
+ list_add_rcu(&iomap_ops.list, &iomap_head);
+}
+EXPORT_SYMBOL(nfit_test_setup);
+
+void nfit_test_teardown(void)
+{
+ list_del_rcu(&iomap_ops.list);
+ synchronize_rcu();
+}
+EXPORT_SYMBOL(nfit_test_teardown);
+
+static struct nfit_test_resource *get_nfit_res(resource_size_t resource)
+{
+ struct iomap_ops *ops;
+
+ ops = list_first_or_null_rcu(&iomap_head, typeof(*ops), list);
+ if (ops)
+ return ops->nfit_test_lookup(resource);
+ return NULL;
+}
+
+void __iomem *__nfit_test_ioremap(resource_size_t offset, unsigned long size,
+ void __iomem *(*fallback_fn)(resource_size_t, unsigned long))
+{
+ struct nfit_test_resource *nfit_res;
+
+ rcu_read_lock();
+ nfit_res = get_nfit_res(offset);
+ rcu_read_unlock();
+ if (nfit_res)
+ return (void __iomem *) nfit_res->buf + offset
+ - nfit_res->res->start;
+ return fallback_fn(offset, size);
+}
+
+void __iomem *__wrap_ioremap_cache(resource_size_t offset, unsigned long size)
+{
+ return __nfit_test_ioremap(offset, size, ioremap_cache);
+}
+EXPORT_SYMBOL(__wrap_ioremap_cache);
+
+void __iomem *__wrap_ioremap_nocache(resource_size_t offset, unsigned long size)
+{
+ return __nfit_test_ioremap(offset, size, ioremap_nocache);
+}
+EXPORT_SYMBOL(__wrap_ioremap_nocache);
+
+void __wrap_iounmap(volatile void __iomem *addr)
+{
+ struct nfit_test_resource *nfit_res;
+
+ rcu_read_lock();
+ nfit_res = get_nfit_res((unsigned long) addr);
+ rcu_read_unlock();
+ if (nfit_res)
+ return;
+ return iounmap(addr);
+}
+EXPORT_SYMBOL(__wrap_iounmap);
+
+struct resource *__wrap___request_region(struct resource *parent,
+ resource_size_t start, resource_size_t n, const char *name,
+ int flags)
+{
+ struct nfit_test_resource *nfit_res;
+
+ if (parent == &iomem_resource) {
+ rcu_read_lock();
+ nfit_res = get_nfit_res(start);
+ rcu_read_unlock();
+ if (nfit_res) {
+ struct resource *res = nfit_res->res + 1;
+
+ if (start + n > nfit_res->res->start
+ + resource_size(nfit_res->res)) {
+ pr_debug("%s: start: %llx n: %llx overflow: %pr\n",
+ __func__, start, n,
+ nfit_res->res);
+ return NULL;
+ }
+
+ res->start = start;
+ res->end = start + n - 1;
+ res->name = name;
+ res->flags = resource_type(parent);
+ res->flags |= IORESOURCE_BUSY | flags;
+ pr_debug("%s: %pr\n", __func__, res);
+ return res;
+ }
+ }
+ return __request_region(parent, start, n, name, flags);
+}
+EXPORT_SYMBOL(__wrap___request_region);
+
+void __wrap___release_region(struct resource *parent, resource_size_t start,
+ resource_size_t n)
+{
+ struct nfit_test_resource *nfit_res;
+
+ if (parent == &iomem_resource) {
+ rcu_read_lock();
+ nfit_res = get_nfit_res(start);
+ rcu_read_unlock();
+ if (nfit_res) {
+ struct resource *res = nfit_res->res + 1;
+
+ if (start != res->start || resource_size(res) != n)
+ pr_info("%s: start: %llx n: %llx mismatch: %pr\n",
+ __func__, start, n, res);
+ else
+ memset(res, 0, sizeof(*res));
+ return;
+ }
+ }
+ __release_region(parent, start, n);
+}
+EXPORT_SYMBOL(__wrap___release_region);
+
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/block/nd/test/nfit.c b/drivers/block/nd/test/nfit.c
new file mode 100644
index 000000000000..8691a903515b
--- /dev/null
+++ b/drivers/block/nd/test/nfit.c
@@ -0,0 +1,1025 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+#include <linux/platform_device.h>
+#include <linux/dma-mapping.h>
+#include <linux/module.h>
+#include <linux/sizes.h>
+#include <linux/slab.h>
+#include "nfit_test.h"
+
+#include "../acpi_nfit.h"
+#include "../libnd.h"
+
+/*
+ * Generate an NFIT table to describe the following topology:
+ *
+ * BUS0: Interleaved PMEM regions, and aliasing with BLK regions
+ *
+ * (a) (b) DIMM BLK-REGION
+ * +----------+--------------+----------+---------+
+ * +------+ | blk2.0 | pm0.0 | blk2.1 | pm1.0 | 0 region2
+ * | imc0 +--+- - - - - region0 - - - -+----------+ +
+ * +--+---+ | blk3.0 | pm0.0 | blk3.1 | pm1.0 | 1 region3
+ * | +----------+--------------v----------v v
+ * +--+---+ | |
+ * | cpu0 | region1
+ * +--+---+ | |
+ * | +-------------------------^----------^ ^
+ * +--+---+ | blk4.0 | pm1.0 | 2 region4
+ * | imc1 +--+-------------------------+----------+ +
+ * +------+ | blk5.0 | pm1.0 | 3 region5
+ * +-------------------------+----------+-+-------+
+ *
+ * *) In this layout we have four dimms and two memory controllers in one
+ * socket. Each unique interface (BLK or PMEM) to DPA space
+ * is identified by a region device with a dynamically assigned id.
+ *
+ * *) The first portion of dimm0 and dimm1 are interleaved as REGION0.
+ * A single PMEM namespace "pm0.0" is created using half of the
+ * REGION0 SPA-range. REGION0 spans dimm0 and dimm1. PMEM namespace
+ * allocate from from the bottom of a region. The unallocated
+ * portion of REGION0 aliases with REGION2 and REGION3. That
+ * unallacted capacity is reclaimed as BLK namespaces ("blk2.0" and
+ * "blk3.0") starting at the base of each DIMM to offset (a) in those
+ * DIMMs. "pm0.0", "blk2.0" and "blk3.0" are free-form readable
+ * names that can be assigned to a namespace.
+ *
+ * *) In the last portion of dimm0 and dimm1 we have an interleaved
+ * SPA range, REGION1, that spans those two dimms as well as dimm2
+ * and dimm3. Some of REGION1 allocated to a PMEM namespace named
+ * "pm1.0" the rest is reclaimed in 4 BLK namespaces (for each
+ * dimm in the interleave set), "blk2.1", "blk3.1", "blk4.0", and
+ * "blk5.0".
+ *
+ * *) The portion of dimm2 and dimm3 that do not participate in the
+ * REGION1 interleaved SPA range (i.e. the DPA address below offset
+ * (b) are also included in the "blk4.0" and "blk5.0" namespaces.
+ * Note, that BLK namespaces need not be contiguous in DPA-space, and
+ * can consume aliased capacity from multiple interleave sets.
+ *
+ * BUS1: Legacy NVDIMM (single contiguous range)
+ *
+ * region2
+ * +---------------------+
+ * |---------------------|
+ * || pm2.0 ||
+ * |---------------------|
+ * +---------------------+
+ *
+ * *) A NFIT-table may describe a simple system-physical-address range
+ * with no BLK aliasing. This type of region may optionally
+ * reference an NVDIMM.
+ */
+enum {
+ NUM_PM = 2,
+ NUM_DCR = 4,
+ NUM_BDW = NUM_DCR,
+ NUM_SPA = NUM_PM + NUM_DCR + NUM_BDW,
+ NUM_MEM = NUM_DCR + NUM_BDW + 2 /* spa0 iset */ + 4 /* spa1 iset */,
+ DIMM_SIZE = SZ_32M,
+ LABEL_SIZE = SZ_128K,
+ SPA0_SIZE = DIMM_SIZE,
+ SPA1_SIZE = DIMM_SIZE*2,
+ SPA2_SIZE = DIMM_SIZE,
+ BDW_SIZE = 64 << 8,
+ DCR_SIZE = 12,
+ NUM_NFITS = 2, /* permit testing multiple NFITs per system */
+};
+
+struct nfit_test_dcr {
+ __le64 bdw_addr;
+ __le32 bdw_status;
+ __u8 aperature[BDW_SIZE];
+};
+
+static u32 handle[NUM_DCR] = {
+ [0] = NFIT_DIMM_HANDLE(0, 0, 0, 0, 0),
+ [1] = NFIT_DIMM_HANDLE(0, 0, 0, 0, 1),
+ [2] = NFIT_DIMM_HANDLE(0, 0, 1, 0, 0),
+ [3] = NFIT_DIMM_HANDLE(0, 0, 1, 0, 1),
+};
+
+struct nfit_test {
+ struct acpi_nfit_desc acpi_desc;
+ struct platform_device pdev;
+ struct list_head resources;
+ void *nfit_buf;
+ dma_addr_t nfit_dma;
+ size_t nfit_size;
+ int num_dcr;
+ int num_pm;
+ void **dimm;
+ dma_addr_t *dimm_dma;
+ void **label;
+ dma_addr_t *label_dma;
+ void **spa_set;
+ dma_addr_t *spa_set_dma;
+ struct nfit_test_dcr **dcr;
+ dma_addr_t *dcr_dma;
+ int (*alloc)(struct nfit_test *t);
+ void (*setup)(struct nfit_test *t);
+};
+
+static struct nfit_test *to_nfit_test(struct device *dev)
+{
+ struct platform_device *pdev = to_platform_device(dev);
+
+ return container_of(pdev, struct nfit_test, pdev);
+}
+
+static int nfit_test_ctl(struct nd_bus_descriptor *nd_desc,
+ struct nd_dimm *nd_dimm, unsigned int cmd, void *buf,
+ unsigned int buf_len)
+{
+ return -ENOTTY;
+}
+
+static DEFINE_SPINLOCK(nfit_test_lock);
+static struct nfit_test *instances[NUM_NFITS];
+
+static void release_nfit_res(void *data)
+{
+ struct nfit_test_resource *nfit_res = data;
+ struct resource *res = nfit_res->res;
+
+ spin_lock(&nfit_test_lock);
+ list_del(&nfit_res->list);
+ spin_unlock(&nfit_test_lock);
+
+ if (is_vmalloc_addr(nfit_res->buf))
+ vfree(nfit_res->buf);
+ else
+ dma_free_coherent(nfit_res->dev, resource_size(res),
+ nfit_res->buf, res->start);
+ kfree(res);
+ kfree(nfit_res);
+}
+
+static void *__test_alloc(struct nfit_test *t, size_t size, dma_addr_t *dma,
+ void *buf)
+{
+ struct device *dev = &t->pdev.dev;
+ struct resource *res = kzalloc(sizeof(*res) * 2, GFP_KERNEL);
+ struct nfit_test_resource *nfit_res = kzalloc(sizeof(*nfit_res),
+ GFP_KERNEL);
+ int rc;
+
+ if (!res || !buf || !nfit_res)
+ goto err;
+ rc = devm_add_action(dev, release_nfit_res, nfit_res);
+ if (rc)
+ goto err;
+ INIT_LIST_HEAD(&nfit_res->list);
+ memset(buf, 0, size);
+ nfit_res->dev = dev;
+ nfit_res->buf = buf;
+ nfit_res->res = res;
+ res->start = *dma;
+ res->end = *dma + size - 1;
+ res->name = "NFIT";
+ spin_lock(&nfit_test_lock);
+ list_add(&nfit_res->list, &t->resources);
+ spin_unlock(&nfit_test_lock);
+
+ return nfit_res->buf;
+ err:
+ if (buf && !is_vmalloc_addr(buf))
+ dma_free_coherent(dev, size, buf, *dma);
+ else if (buf)
+ vfree(buf);
+ kfree(res);
+ kfree(nfit_res);
+ return NULL;
+}
+
+static void *test_alloc(struct nfit_test *t, size_t size, dma_addr_t *dma)
+{
+ void *buf = vmalloc(size);
+
+ *dma = (unsigned long) buf;
+ return __test_alloc(t, size, dma, buf);
+}
+
+static void *test_alloc_coherent(struct nfit_test *t, size_t size, dma_addr_t *dma)
+{
+ struct device *dev = &t->pdev.dev;
+ void *buf = dma_alloc_coherent(dev, size, dma, GFP_KERNEL);
+
+ return __test_alloc(t, size, dma, buf);
+}
+
+static struct nfit_test_resource *nfit_test_lookup(resource_size_t addr)
+{
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(instances); i++) {
+ struct nfit_test_resource *n, *nfit_res = NULL;
+ struct nfit_test *t = instances[i];
+
+ if (!t)
+ continue;
+ spin_lock(&nfit_test_lock);
+ list_for_each_entry(n, &t->resources, list) {
+ if (addr >= n->res->start && (addr < n->res->start
+ + resource_size(n->res))) {
+ nfit_res = n;
+ break;
+ } else if (addr >= (unsigned long) n->buf
+ && (addr < (unsigned long) n->buf
+ + resource_size(n->res))) {
+ nfit_res = n;
+ break;
+ }
+ }
+ spin_unlock(&nfit_test_lock);
+ if (nfit_res)
+ return nfit_res;
+ }
+
+ return NULL;
+}
+
+static int nfit_test0_alloc(struct nfit_test *t)
+{
+ size_t nfit_size = sizeof(struct acpi_nfit)
+ + sizeof(struct acpi_nfit_spa) * NUM_SPA
+ + sizeof(struct acpi_nfit_memdev) * NUM_MEM
+ + sizeof(struct acpi_nfit_dcr) * NUM_DCR
+ + sizeof(struct acpi_nfit_bdw) * NUM_BDW;
+ int i;
+
+ t->nfit_buf = test_alloc(t, nfit_size, &t->nfit_dma);
+ if (!t->nfit_buf)
+ return -ENOMEM;
+ t->nfit_size = nfit_size;
+
+ t->spa_set[0] = test_alloc_coherent(t, SPA0_SIZE, &t->spa_set_dma[0]);
+ if (!t->spa_set[0])
+ return -ENOMEM;
+
+ t->spa_set[1] = test_alloc_coherent(t, SPA1_SIZE, &t->spa_set_dma[1]);
+ if (!t->spa_set[1])
+ return -ENOMEM;
+
+ for (i = 0; i < NUM_DCR; i++) {
+ t->dimm[i] = test_alloc(t, DIMM_SIZE, &t->dimm_dma[i]);
+ if (!t->dimm[i])
+ return -ENOMEM;
+
+ t->label[i] = test_alloc(t, LABEL_SIZE, &t->label_dma[i]);
+ if (!t->label[i])
+ return -ENOMEM;
+ }
+
+ for (i = 0; i < NUM_DCR; i++) {
+ t->dcr[i] = test_alloc(t, LABEL_SIZE, &t->dcr_dma[i]);
+ if (!t->dcr[i])
+ return -ENOMEM;
+ }
+
+ return 0;
+}
+
+static u8 nfit_checksum(void *buf, size_t size)
+{
+ u8 sum, *data = buf;
+ size_t i;
+
+ for (sum = 0, i = 0; i < size; i++)
+ sum += data[i];
+ return 0 - sum;
+}
+
+static int nfit_test1_alloc(struct nfit_test *t)
+{
+ size_t nfit_size = sizeof(struct acpi_nfit)
+ + sizeof(struct acpi_nfit_spa) + sizeof(struct acpi_nfit_memdev)
+ + sizeof(struct acpi_nfit_dcr);
+
+ t->nfit_buf = test_alloc(t, nfit_size, &t->nfit_dma);
+ if (!t->nfit_buf)
+ return -ENOMEM;
+ t->nfit_size = nfit_size;
+
+ t->spa_set[0] = test_alloc_coherent(t, SPA2_SIZE, &t->spa_set_dma[0]);
+ if (!t->spa_set[0])
+ return -ENOMEM;
+
+ return 0;
+}
+
+static void nfit_test0_setup(struct nfit_test *t)
+{
+ struct nd_bus_descriptor *nd_desc;
+ struct acpi_nfit_memdev *memdev;
+ void *nfit_buf = t->nfit_buf;
+ size_t size = t->nfit_size;
+ struct acpi_nfit_spa *spa;
+ struct acpi_nfit_dcr *dcr;
+ struct acpi_nfit_bdw *bdw;
+ struct acpi_nfit *nfit;
+ unsigned int offset;
+
+ /* nfit header */
+ nfit = nfit_buf;
+ memcpy(nfit->signature, "NFIT", 4);
+ nfit->length = size;
+ nfit->revision = 1;
+ memcpy(nfit->oemid, "NDTEST", 6);
+ nfit->oem_tbl_id = 0x1234;
+ nfit->oem_revision = 1;
+ nfit->creator_id = 0xabcd0000;
+ nfit->creator_revision = 1;
+
+ /*
+ * spa0 (interleave first half of dimm0 and dimm1, note storage
+ * does not actually alias the related block-data-window
+ * regions)
+ */
+ spa = nfit_buf + sizeof(*nfit);
+ spa->type = NFIT_TABLE_SPA;
+ spa->length = sizeof(*spa);
+ memcpy(spa->type_uuid, &nfit_spa_uuid_pm, 16);
+ spa->spa_index = 0+1;
+ spa->spa_base = t->spa_set_dma[0];
+ spa->spa_length = SPA0_SIZE;
+
+ /*
+ * spa1 (interleave last half of the 4 DIMMS, note storage
+ * does not actually alias the related block-data-window
+ * regions)
+ */
+ spa = nfit_buf + sizeof(*nfit) + sizeof(*spa);
+ spa->type = NFIT_TABLE_SPA;
+ spa->length = sizeof(*spa);
+ memcpy(spa->type_uuid, &nfit_spa_uuid_pm, 16);
+ spa->spa_index = 1+1;
+ spa->spa_base = t->spa_set_dma[1];
+ spa->spa_length = SPA1_SIZE;
+
+ /* spa2 (dcr0) dimm0 */
+ spa = nfit_buf + sizeof(*nfit) + sizeof(*spa) * 2;
+ spa->type = NFIT_TABLE_SPA;
+ spa->length = sizeof(*spa);
+ memcpy(spa->type_uuid, &nfit_spa_uuid_dcr, 16);
+ spa->spa_index = 2+1;
+ spa->spa_base = t->dcr_dma[0];
+ spa->spa_length = DCR_SIZE;
+
+ /* spa3 (dcr1) dimm1 */
+ spa = nfit_buf + sizeof(*nfit) + sizeof(*spa) * 3;
+ spa->type = NFIT_TABLE_SPA;
+ spa->length = sizeof(*spa);
+ memcpy(spa->type_uuid, &nfit_spa_uuid_dcr, 16);
+ spa->spa_index = 3+1;
+ spa->spa_base = t->dcr_dma[1];
+ spa->spa_length = DCR_SIZE;
+
+ /* spa4 (dcr2) dimm2 */
+ spa = nfit_buf + sizeof(*nfit) + sizeof(*spa) * 4;
+ spa->type = NFIT_TABLE_SPA;
+ spa->length = sizeof(*spa);
+ memcpy(spa->type_uuid, &nfit_spa_uuid_dcr, 16);
+ spa->spa_index = 4+1;
+ spa->spa_base = t->dcr_dma[2];
+ spa->spa_length = DCR_SIZE;
+
+ /* spa5 (dcr3) dimm3 */
+ spa = nfit_buf + sizeof(*nfit) + sizeof(*spa) * 5;
+ spa->type = NFIT_TABLE_SPA;
+ spa->length = sizeof(*spa);
+ memcpy(spa->type_uuid, &nfit_spa_uuid_dcr, 16);
+ spa->spa_index = 5+1;
+ spa->spa_base = t->dcr_dma[3];
+ spa->spa_length = DCR_SIZE;
+
+ /* spa6 (bdw for dcr0) dimm0 */
+ spa = nfit_buf + sizeof(*nfit) + sizeof(*spa) * 6;
+ spa->type = NFIT_TABLE_SPA;
+ spa->length = sizeof(*spa);
+ memcpy(spa->type_uuid, &nfit_spa_uuid_bdw, 16);
+ spa->spa_index = 6+1;
+ spa->spa_base = t->dimm_dma[0];
+ spa->spa_length = DIMM_SIZE;
+
+ /* spa7 (bdw for dcr1) dimm1 */
+ spa = nfit_buf + sizeof(*nfit) + sizeof(*spa) * 7;
+ spa->type = NFIT_TABLE_SPA;
+ spa->length = sizeof(*spa);
+ memcpy(spa->type_uuid, &nfit_spa_uuid_bdw, 16);
+ spa->spa_index = 7+1;
+ spa->spa_base = t->dimm_dma[1];
+ spa->spa_length = DIMM_SIZE;
+
+ /* spa8 (bdw for dcr2) dimm2 */
+ spa = nfit_buf + sizeof(*nfit) + sizeof(*spa) * 8;
+ spa->type = NFIT_TABLE_SPA;
+ spa->length = sizeof(*spa);
+ memcpy(spa->type_uuid, &nfit_spa_uuid_bdw, 16);
+ spa->spa_index = 8+1;
+ spa->spa_base = t->dimm_dma[2];
+ spa->spa_length = DIMM_SIZE;
+
+ /* spa9 (bdw for dcr3) dimm3 */
+ spa = nfit_buf + sizeof(*nfit) + sizeof(*spa) * 9;
+ spa->type = NFIT_TABLE_SPA;
+ spa->length = sizeof(*spa);
+ memcpy(spa->type_uuid, &nfit_spa_uuid_bdw, 16);
+ spa->spa_index = 9+1;
+ spa->spa_base = t->dimm_dma[3];
+ spa->spa_length = DIMM_SIZE;
+
+ offset = sizeof(*nfit) + sizeof(*spa) * 10;
+ /* mem-region0 (spa0, dimm0) */
+ memdev = nfit_buf + offset;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[0];
+ memdev->phys_id = 0;
+ memdev->region_id = 0;
+ memdev->spa_index = 0+1;
+ memdev->dcr_index = 0+1;
+ memdev->region_len = SPA0_SIZE/2;
+ memdev->region_spa_offset = t->spa_set_dma[0];
+ memdev->region_dpa = 0;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 2;
+
+ /* mem-region1 (spa0, dimm1) */
+ memdev = nfit_buf + offset + sizeof(struct acpi_nfit_memdev);
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[1];
+ memdev->phys_id = 1;
+ memdev->region_id = 0;
+ memdev->spa_index = 0+1;
+ memdev->dcr_index = 1+1;
+ memdev->region_len = SPA0_SIZE/2;
+ memdev->region_spa_offset = t->spa_set_dma[0] + SPA0_SIZE/2;
+ memdev->region_dpa = 0;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 2;
+
+ /* mem-region2 (spa1, dimm0) */
+ memdev = nfit_buf + offset + sizeof(struct acpi_nfit_memdev) * 2;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[0];
+ memdev->phys_id = 0;
+ memdev->region_id = 1;
+ memdev->spa_index = 1+1;
+ memdev->dcr_index = 0+1;
+ memdev->region_len = SPA1_SIZE/4;
+ memdev->region_spa_offset = t->spa_set_dma[1];
+ memdev->region_dpa = SPA0_SIZE/2;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 4;
+
+ /* mem-region3 (spa1, dimm1) */
+ memdev = nfit_buf + offset + sizeof(struct acpi_nfit_memdev) * 3;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[1];
+ memdev->phys_id = 1;
+ memdev->region_id = 1;
+ memdev->spa_index = 1+1;
+ memdev->dcr_index = 1+1;
+ memdev->region_len = SPA1_SIZE/4;
+ memdev->region_spa_offset = t->spa_set_dma[1] + SPA1_SIZE/4;
+ memdev->region_dpa = SPA0_SIZE/2;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 4;
+
+ /* mem-region4 (spa1, dimm2) */
+ memdev = nfit_buf + offset + sizeof(struct acpi_nfit_memdev) * 4;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[2];
+ memdev->phys_id = 2;
+ memdev->region_id = 0;
+ memdev->spa_index = 1+1;
+ memdev->dcr_index = 2+1;
+ memdev->region_len = SPA1_SIZE/4;
+ memdev->region_spa_offset = t->spa_set_dma[1] + 2*SPA1_SIZE/4;
+ memdev->region_dpa = SPA0_SIZE/2;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 4;
+
+ /* mem-region5 (spa1, dimm3) */
+ memdev = nfit_buf + offset + sizeof(struct acpi_nfit_memdev) * 5;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[3];
+ memdev->phys_id = 3;
+ memdev->region_id = 0;
+ memdev->spa_index = 1+1;
+ memdev->dcr_index = 3+1;
+ memdev->region_len = SPA1_SIZE/4;
+ memdev->region_spa_offset = t->spa_set_dma[1] + 3*SPA1_SIZE/4;
+ memdev->region_dpa = SPA0_SIZE/2;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 4;
+
+ /* mem-region6 (spa/dcr0, dimm0) */
+ memdev = nfit_buf + offset + sizeof(struct acpi_nfit_memdev) * 6;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[0];
+ memdev->phys_id = 0;
+ memdev->region_id = 0;
+ memdev->spa_index = 2+1;
+ memdev->dcr_index = 0+1;
+ memdev->region_len = 0;
+ memdev->region_spa_offset = 0;
+ memdev->region_dpa = 0;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 1;
+
+ /* mem-region7 (spa/dcr1, dimm1) */
+ memdev = nfit_buf + offset + sizeof(struct acpi_nfit_memdev) * 7;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[1];
+ memdev->phys_id = 1;
+ memdev->region_id = 0;
+ memdev->spa_index = 3+1;
+ memdev->dcr_index = 1+1;
+ memdev->region_len = 0;
+ memdev->region_spa_offset = 0;
+ memdev->region_dpa = 0;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 1;
+
+ /* mem-region8 (spa/dcr2, dimm2) */
+ memdev = nfit_buf + offset + sizeof(struct acpi_nfit_memdev) * 8;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[2];
+ memdev->phys_id = 2;
+ memdev->region_id = 0;
+ memdev->spa_index = 4+1;
+ memdev->dcr_index = 2+1;
+ memdev->region_len = 0;
+ memdev->region_spa_offset = 0;
+ memdev->region_dpa = 0;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 1;
+
+ /* mem-region9 (spa/dcr3, dimm3) */
+ memdev = nfit_buf + offset + sizeof(struct acpi_nfit_memdev) * 9;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[3];
+ memdev->phys_id = 3;
+ memdev->region_id = 0;
+ memdev->spa_index = 5+1;
+ memdev->dcr_index = 3+1;
+ memdev->region_len = 0;
+ memdev->region_spa_offset = 0;
+ memdev->region_dpa = 0;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 1;
+
+ /* mem-region10 (spa/bdw0, dimm0) */
+ memdev = nfit_buf + offset + sizeof(struct acpi_nfit_memdev) * 10;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[0];
+ memdev->phys_id = 0;
+ memdev->region_id = 0;
+ memdev->spa_index = 6+1;
+ memdev->dcr_index = 0+1;
+ memdev->region_len = 0;
+ memdev->region_spa_offset = 0;
+ memdev->region_dpa = 0;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 1;
+
+ /* mem-region11 (spa/bdw1, dimm1) */
+ memdev = nfit_buf + offset + sizeof(struct acpi_nfit_memdev) * 11;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[1];
+ memdev->phys_id = 1;
+ memdev->region_id = 0;
+ memdev->spa_index = 7+1;
+ memdev->dcr_index = 1+1;
+ memdev->region_len = 0;
+ memdev->region_spa_offset = 0;
+ memdev->region_dpa = 0;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 1;
+
+ /* mem-region12 (spa/bdw2, dimm2) */
+ memdev = nfit_buf + offset + sizeof(struct acpi_nfit_memdev) * 12;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[2];
+ memdev->phys_id = 2;
+ memdev->region_id = 0;
+ memdev->spa_index = 8+1;
+ memdev->dcr_index = 2+1;
+ memdev->region_len = 0;
+ memdev->region_spa_offset = 0;
+ memdev->region_dpa = 0;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 1;
+
+ /* mem-region13 (spa/dcr3, dimm3) */
+ memdev = nfit_buf + offset + sizeof(struct acpi_nfit_memdev) * 13;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = handle[3];
+ memdev->phys_id = 3;
+ memdev->region_id = 0;
+ memdev->spa_index = 9+1;
+ memdev->dcr_index = 3+1;
+ memdev->region_len = 0;
+ memdev->region_spa_offset = 0;
+ memdev->region_dpa = 0;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 1;
+
+ offset = offset + sizeof(struct acpi_nfit_memdev) * 14;
+ /* dcr-descriptor0 */
+ dcr = nfit_buf + offset;
+ dcr->type = NFIT_TABLE_DCR;
+ dcr->length = sizeof(struct acpi_nfit_dcr);
+ dcr->dcr_index = 0+1;
+ dcr->vendor_id = 0xabcd;
+ dcr->device_id = 0;
+ dcr->revision_id = 1;
+ dcr->serial_number = ~handle[0];
+ dcr->num_bcw = 1;
+ dcr->bcw_size = DCR_SIZE;
+ dcr->cmd_offset = 0;
+ dcr->cmd_size = 8;
+ dcr->status_offset = 8;
+ dcr->status_size = 4;
+
+ /* dcr-descriptor1 */
+ dcr = nfit_buf + offset + sizeof(struct acpi_nfit_dcr);
+ dcr->type = NFIT_TABLE_DCR;
+ dcr->length = sizeof(struct acpi_nfit_dcr);
+ dcr->dcr_index = 1+1;
+ dcr->vendor_id = 0xabcd;
+ dcr->device_id = 0;
+ dcr->revision_id = 1;
+ dcr->serial_number = ~handle[1];
+ dcr->num_bcw = 1;
+ dcr->bcw_size = DCR_SIZE;
+ dcr->cmd_offset = 0;
+ dcr->cmd_size = 8;
+ dcr->status_offset = 8;
+ dcr->status_size = 4;
+
+ /* dcr-descriptor2 */
+ dcr = nfit_buf + offset + sizeof(struct acpi_nfit_dcr) * 2;
+ dcr->type = NFIT_TABLE_DCR;
+ dcr->length = sizeof(struct acpi_nfit_dcr);
+ dcr->dcr_index = 2+1;
+ dcr->vendor_id = 0xabcd;
+ dcr->device_id = 0;
+ dcr->revision_id = 1;
+ dcr->serial_number = ~handle[2];
+ dcr->num_bcw = 1;
+ dcr->bcw_size = DCR_SIZE;
+ dcr->cmd_offset = 0;
+ dcr->cmd_size = 8;
+ dcr->status_offset = 8;
+ dcr->status_size = 4;
+
+ /* dcr-descriptor3 */
+ dcr = nfit_buf + offset + sizeof(struct acpi_nfit_dcr) * 3;
+ dcr->type = NFIT_TABLE_DCR;
+ dcr->length = sizeof(struct acpi_nfit_dcr);
+ dcr->dcr_index = 3+1;
+ dcr->vendor_id = 0xabcd;
+ dcr->device_id = 0;
+ dcr->revision_id = 1;
+ dcr->serial_number = ~handle[3];
+ dcr->num_bcw = 1;
+ dcr->bcw_size = DCR_SIZE;
+ dcr->cmd_offset = 0;
+ dcr->cmd_size = 8;
+ dcr->status_offset = 8;
+ dcr->status_size = 4;
+
+ offset = offset + sizeof(struct acpi_nfit_dcr) * 4;
+ /* bdw0 (spa/dcr0, dimm0) */
+ bdw = nfit_buf + offset;
+ bdw->type = NFIT_TABLE_BDW;
+ bdw->length = sizeof(struct acpi_nfit_bdw);
+ bdw->dcr_index = 0+1;
+ bdw->num_bdw = 1;
+ bdw->bdw_offset = 0;
+ bdw->bdw_size = BDW_SIZE;
+ bdw->blk_capacity = DIMM_SIZE;
+ bdw->blk_offset = 0;
+
+ /* bdw1 (spa/dcr1, dimm1) */
+ bdw = nfit_buf + offset + sizeof(struct acpi_nfit_bdw);
+ bdw->type = NFIT_TABLE_BDW;
+ bdw->length = sizeof(struct acpi_nfit_bdw);
+ bdw->dcr_index = 1+1;
+ bdw->num_bdw = 1;
+ bdw->bdw_offset = 0;
+ bdw->bdw_size = BDW_SIZE;
+ bdw->blk_capacity = DIMM_SIZE;
+ bdw->blk_offset = 0;
+
+ /* bdw2 (spa/dcr2, dimm2) */
+ bdw = nfit_buf + offset + sizeof(struct acpi_nfit_bdw) * 2;
+ bdw->type = NFIT_TABLE_BDW;
+ bdw->length = sizeof(struct acpi_nfit_bdw);
+ bdw->dcr_index = 2+1;
+ bdw->num_bdw = 1;
+ bdw->bdw_offset = 0;
+ bdw->bdw_size = BDW_SIZE;
+ bdw->blk_capacity = DIMM_SIZE;
+ bdw->blk_offset = 0;
+
+ /* bdw3 (spa/dcr3, dimm3) */
+ bdw = nfit_buf + offset + sizeof(struct acpi_nfit_bdw) * 3;
+ bdw->type = NFIT_TABLE_BDW;
+ bdw->length = sizeof(struct acpi_nfit_bdw);
+ bdw->dcr_index = 3+1;
+ bdw->num_bdw = 1;
+ bdw->bdw_offset = 0;
+ bdw->bdw_size = BDW_SIZE;
+ bdw->blk_capacity = DIMM_SIZE;
+ bdw->blk_offset = 0;
+
+ nfit->checksum = nfit_checksum(nfit_buf, size);
+
+ nd_desc = &t->acpi_desc.nd_desc;
+ nd_desc->ndctl = nfit_test_ctl;
+}
+
+static void nfit_test1_setup(struct nfit_test *t)
+{
+ size_t size = t->nfit_size, offset;
+ void *nfit_buf = t->nfit_buf;
+ struct acpi_nfit_memdev *memdev;
+ struct acpi_nfit_dcr *dcr;
+ struct acpi_nfit_spa *spa;
+ struct acpi_nfit *nfit;
+
+ /* nfit header */
+ nfit = nfit_buf;
+ memcpy(nfit->signature, "NFIT", 4);
+ nfit->length = size;
+ nfit->revision = 1;
+ memcpy(nfit->oemid, "NDTEST", 6);
+ nfit->oem_tbl_id = 0x1234;
+ nfit->oem_revision = 1;
+ nfit->creator_id = 0xabcd0000;
+ nfit->creator_revision = 1;
+
+ offset = sizeof(*nfit);
+ /* spa0 (flat range with no bdw aliasing) */
+ spa = nfit_buf + offset;
+ spa->type = NFIT_TABLE_SPA;
+ spa->length = sizeof(*spa);
+ memcpy(spa->type_uuid, &nfit_spa_uuid_pm, 16);
+ spa->spa_index = 0+1;
+ spa->spa_base = t->spa_set_dma[0];
+ spa->spa_length = SPA2_SIZE;
+
+ offset += sizeof(*spa);
+ /* mem-region0 (spa0, dimm0) */
+ memdev = nfit_buf + offset;
+ memdev->type = NFIT_TABLE_MEM;
+ memdev->length = sizeof(*memdev);
+ memdev->nfit_handle = 0;
+ memdev->phys_id = 0;
+ memdev->region_id = 0;
+ memdev->spa_index = 0+1;
+ memdev->dcr_index = 0+1;
+ memdev->region_len = SPA2_SIZE;
+ memdev->region_spa_offset = 0;
+ memdev->region_dpa = 0;
+ memdev->idt_index = 0;
+ memdev->interleave_ways = 1;
+
+ offset += sizeof(*memdev);
+ /* dcr-descriptor0 */
+ dcr = nfit_buf + offset;
+ dcr->type = NFIT_TABLE_DCR;
+ dcr->length = sizeof(struct acpi_nfit_dcr);
+ dcr->dcr_index = 0+1;
+ dcr->vendor_id = 0xabcd;
+ dcr->device_id = 0;
+ dcr->revision_id = 1;
+ dcr->serial_number = ~0;
+ dcr->num_bcw = 0;
+ dcr->bcw_size = 0;
+ dcr->cmd_offset = 0;
+ dcr->cmd_size = 0;
+ dcr->status_offset = 0;
+ dcr->status_size = 0;
+
+ nfit->checksum = nfit_checksum(nfit_buf, size);
+}
+
+static int nfit_test_probe(struct platform_device *pdev)
+{
+ struct nd_bus_descriptor *nd_desc;
+ struct acpi_nfit_desc *acpi_desc;
+ struct device *dev = &pdev->dev;
+ struct nfit_test *nfit_test;
+ int rc;
+
+ nfit_test = to_nfit_test(&pdev->dev);
+
+ /* common alloc */
+ if (nfit_test->num_dcr) {
+ int num = nfit_test->num_dcr;
+
+ nfit_test->dimm = devm_kcalloc(dev, num, sizeof(void *), GFP_KERNEL);
+ nfit_test->dimm_dma = devm_kcalloc(dev, num, sizeof(dma_addr_t), GFP_KERNEL);
+ nfit_test->label = devm_kcalloc(dev, num, sizeof(void *), GFP_KERNEL);
+ nfit_test->label_dma = devm_kcalloc(dev, num, sizeof(dma_addr_t), GFP_KERNEL);
+ nfit_test->dcr = devm_kcalloc(dev, num, sizeof(struct nfit_test_dcr *), GFP_KERNEL);
+ nfit_test->dcr_dma = devm_kcalloc(dev, num, sizeof(dma_addr_t), GFP_KERNEL);
+ if (nfit_test->dimm && nfit_test->dimm_dma && nfit_test->label
+ && nfit_test->label_dma && nfit_test->dcr
+ && nfit_test->dcr_dma)
+ /* pass */;
+ else
+ return -ENOMEM;
+ }
+
+ if (nfit_test->num_pm) {
+ int num = nfit_test->num_pm;
+
+ nfit_test->spa_set = devm_kcalloc(dev, num, sizeof(void *), GFP_KERNEL);
+ nfit_test->spa_set_dma = devm_kcalloc(dev, num,
+ sizeof(dma_addr_t), GFP_KERNEL);
+ if (nfit_test->spa_set && nfit_test->spa_set_dma)
+ /* pass */;
+ else
+ return -ENOMEM;
+ }
+
+ /* per-nfit specific alloc */
+ if (nfit_test->alloc(nfit_test))
+ return -ENOMEM;
+
+ nfit_test->setup(nfit_test);
+ acpi_desc = &nfit_test->acpi_desc;
+ acpi_desc->dev = &pdev->dev;
+ acpi_desc->nfit = nfit_test->nfit_buf;
+ nd_desc = &acpi_desc->nd_desc;
+ acpi_desc->nd_bus = nd_bus_register(&pdev->dev, nd_desc);
+ if (!acpi_desc->nd_bus)
+ return -ENXIO;
+
+ rc = nd_acpi_nfit_init(acpi_desc, nfit_test->nfit_size);
+ if (rc) {
+ nd_bus_unregister(acpi_desc->nd_bus);
+ return rc;
+ }
+
+ return 0;
+}
+
+static int nfit_test_remove(struct platform_device *pdev)
+{
+ struct nfit_test *nfit_test = to_nfit_test(&pdev->dev);
+ struct acpi_nfit_desc *acpi_desc = &nfit_test->acpi_desc;
+
+ nd_bus_unregister(acpi_desc->nd_bus);
+
+ return 0;
+}
+
+static void nfit_test_release(struct device *dev)
+{
+ struct nfit_test *nfit_test = to_nfit_test(dev);
+
+ kfree(nfit_test);
+}
+
+static const struct platform_device_id nfit_test_id[] = {
+ { KBUILD_MODNAME },
+ { },
+};
+
+static struct platform_driver nfit_test_driver = {
+ .probe = nfit_test_probe,
+ .remove = nfit_test_remove,
+ .driver = {
+ .name = KBUILD_MODNAME,
+ },
+ .id_table = nfit_test_id,
+};
+
+#ifdef CONFIG_CMA_SIZE_MBYTES
+#define CMA_SIZE_MBYTES CONFIG_CMA_SIZE_MBYTES
+#else
+#define CMA_SIZE_MBYTES 0
+#endif
+
+static __init int nfit_test_init(void)
+{
+ int rc, i;
+
+ nfit_test_setup(nfit_test_lookup);
+
+ for (i = 0; i < NUM_NFITS; i++) {
+ struct nfit_test *nfit_test;
+ struct platform_device *pdev;
+ static int once;
+
+ nfit_test = kzalloc(sizeof(*nfit_test), GFP_KERNEL);
+ if (!nfit_test) {
+ rc = -ENOMEM;
+ goto err_register;
+ }
+ INIT_LIST_HEAD(&nfit_test->resources);
+ switch (i) {
+ case 0:
+ nfit_test->num_pm = NUM_PM;
+ nfit_test->num_dcr = NUM_DCR;
+ nfit_test->alloc = nfit_test0_alloc;
+ nfit_test->setup = nfit_test0_setup;
+ break;
+ case 1:
+ nfit_test->num_pm = 1;
+ nfit_test->alloc = nfit_test1_alloc;
+ nfit_test->setup = nfit_test1_setup;
+ break;
+ default:
+ rc = -EINVAL;
+ goto err_register;
+ }
+ pdev = &nfit_test->pdev;
+ pdev->name = KBUILD_MODNAME;
+ pdev->id = i;
+ pdev->dev.release = nfit_test_release;
+ rc = platform_device_register(pdev);
+ if (rc) {
+ put_device(&pdev->dev);
+ goto err_register;
+ }
+
+ rc = dma_coerce_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+ if (rc)
+ goto err_register;
+
+ instances[i] = nfit_test;
+
+ if (!once++) {
+ dma_addr_t dma;
+ void *buf;
+
+ buf = dma_alloc_coherent(&pdev->dev, SZ_128M, &dma,
+ GFP_KERNEL);
+ if (!buf) {
+ rc = -ENOMEM;
+ dev_warn(&pdev->dev, "need 128M of free cma\n");
+ goto err_register;
+ }
+ dma_free_coherent(&pdev->dev, SZ_128M, buf, dma);
+ }
+ }
+
+ rc = platform_driver_register(&nfit_test_driver);
+ if (rc)
+ goto err_register;
+ return 0;
+
+ err_register:
+ for (i = 0; i < NUM_NFITS; i++)
+ if (instances[i])
+ platform_device_unregister(&instances[i]->pdev);
+ nfit_test_teardown();
+ return rc;
+}
+
+static __exit void nfit_test_exit(void)
+{
+ int i;
+
+ platform_driver_unregister(&nfit_test_driver);
+ for (i = 0; i < NUM_NFITS; i++)
+ platform_device_unregister(&instances[i]->pdev);
+ nfit_test_teardown();
+}
+
+module_init(nfit_test_init);
+module_exit(nfit_test_exit);
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Intel Corporation");
diff --git a/drivers/block/nd/test/nfit_test.h b/drivers/block/nd/test/nfit_test.h
new file mode 100644
index 000000000000..7b071478eb94
--- /dev/null
+++ b/drivers/block/nd/test/nfit_test.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#ifndef __NFIT_TEST_H__
+#define __NFIT_TEST_H__
+
+struct nfit_test_resource {
+ struct list_head list;
+ struct resource *res;
+ struct device *dev;
+ void *buf;
+};
+
+typedef struct nfit_test_resource *(*nfit_test_lookup_fn)(resource_size_t);
+void nfit_test_setup(nfit_test_lookup_fn fn);
+void nfit_test_teardown(void);
+#endif
This is the position (device topology) independent method to find all
the libnd buses in the system. The expectation is that there will only
ever be one "nd" bus discovered via /sys/class/nd/ndctl0. However, we
allow for the possibility of multiple buses and they will listed in
discovery order as ndctl0...ndctlN. This character device hosts the
ioctl for passing control messages (inspired by the ACPI-NFIT DSM
interface commands).
Note, nd_ioctl() and the backing ->ndctl() implementation are defined in
a subsequent patch.
Cc: Neil Brown <[email protected]>
Cc: Greg KH <[email protected]>
Cc: <[email protected]>
Cc: Robert Moore <[email protected]>
Cc: Rafael J. Wysocki <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/Makefile | 1
drivers/block/nd/acpi.c | 29 ++++++++++++++
drivers/block/nd/acpi_nfit.h | 5 ++
drivers/block/nd/bus.c | 83 +++++++++++++++++++++++++++++++++++++++
drivers/block/nd/core.c | 87 ++++++++++++++++++++++++++++++++++++++++-
drivers/block/nd/libnd.h | 5 ++
drivers/block/nd/nd-private.h | 6 +++
drivers/block/nd/test/nfit.c | 3 +
8 files changed, 217 insertions(+), 2 deletions(-)
create mode 100644 drivers/block/nd/bus.c
diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
index cf064db92589..7defe18ed009 100644
--- a/drivers/block/nd/Makefile
+++ b/drivers/block/nd/Makefile
@@ -19,3 +19,4 @@ obj-$(CONFIG_NFIT_TEST) += test/
nd_acpi-y := acpi.o
libnd-y := core.o
+libnd-y += bus.o
diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
index 54344ef9c837..dd8505f766ed 100644
--- a/drivers/block/nd/acpi.c
+++ b/drivers/block/nd/acpi.c
@@ -341,6 +341,34 @@ static int nfit_mem_init(struct acpi_nfit_desc *acpi_desc)
return 0;
}
+static ssize_t revision_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_bus *nd_bus = to_nd_bus(dev);
+ struct nd_bus_descriptor *nd_desc = to_nd_desc(nd_bus);
+ struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc);
+
+ return sprintf(buf, "%d\n", acpi_desc->nfit->revision);
+}
+static DEVICE_ATTR_RO(revision);
+
+static struct attribute *nd_acpi_attributes[] = {
+ &dev_attr_revision.attr,
+ NULL,
+};
+
+static struct attribute_group nd_acpi_attribute_group = {
+ .name = "nfit",
+ .attrs = nd_acpi_attributes,
+};
+
+const struct attribute_group *nd_acpi_attribute_groups[] = {
+ &nd_bus_attribute_group,
+ &nd_acpi_attribute_group,
+ NULL,
+};
+EXPORT_SYMBOL_GPL(nd_acpi_attribute_groups);
+
int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
{
struct device *dev = acpi_desc->dev;
@@ -408,6 +436,7 @@ static int nd_acpi_add(struct acpi_device *adev)
nd_desc = &acpi_desc->nd_desc;
nd_desc->provider_name = "ACPI.NFIT";
nd_desc->ndctl = nd_acpi_ctl;
+ nd_desc->attr_groups = nd_acpi_attribute_groups;
acpi_desc->nd_bus = nd_bus_register(dev, nd_desc);
if (!acpi_desc->nd_bus)
diff --git a/drivers/block/nd/acpi_nfit.h b/drivers/block/nd/acpi_nfit.h
index a26f69e32244..b65745ca3cbc 100644
--- a/drivers/block/nd/acpi_nfit.h
+++ b/drivers/block/nd/acpi_nfit.h
@@ -261,5 +261,10 @@ static inline struct acpi_nfit_memdev *__to_nfit_memdev(struct nfit_mem *nfit_me
return nfit_mem->memdev_pmem;
}
+static inline struct acpi_nfit_desc *to_acpi_desc(struct nd_bus_descriptor *nd_desc)
+{
+ return container_of(nd_desc, struct acpi_nfit_desc, nd_desc);
+}
+
int nd_acpi_nfit_init(struct acpi_nfit_desc *nfit, acpi_size sz);
#endif /* __NFIT_H__ */
diff --git a/drivers/block/nd/bus.c b/drivers/block/nd/bus.c
new file mode 100644
index 000000000000..635f2e926426
--- /dev/null
+++ b/drivers/block/nd/bus.c
@@ -0,0 +1,83 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+#include <linux/uaccess.h>
+#include <linux/fcntl.h>
+#include <linux/slab.h>
+#include <linux/fs.h>
+#include <linux/io.h>
+#include "nd-private.h"
+
+static int nd_bus_major;
+static struct class *nd_class;
+
+int nd_bus_create_ndctl(struct nd_bus *nd_bus)
+{
+ dev_t devt = MKDEV(nd_bus_major, nd_bus->id);
+ struct device *dev;
+
+ dev = device_create(nd_class, &nd_bus->dev, devt, nd_bus, "ndctl%d",
+ nd_bus->id);
+
+ if (IS_ERR(dev)) {
+ dev_dbg(&nd_bus->dev, "failed to register ndctl%d: %ld\n",
+ nd_bus->id, PTR_ERR(dev));
+ return PTR_ERR(dev);
+ }
+ return 0;
+}
+
+void nd_bus_destroy_ndctl(struct nd_bus *nd_bus)
+{
+ device_destroy(nd_class, MKDEV(nd_bus_major, nd_bus->id));
+}
+
+static long nd_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+ return -ENXIO;
+}
+
+static const struct file_operations nd_bus_fops = {
+ .owner = THIS_MODULE,
+ .open = nonseekable_open,
+ .unlocked_ioctl = nd_ioctl,
+ .compat_ioctl = nd_ioctl,
+ .llseek = noop_llseek,
+};
+
+int __init nd_bus_init(void)
+{
+ int rc;
+
+ rc = register_chrdev(0, "ndctl", &nd_bus_fops);
+ if (rc < 0)
+ return rc;
+ nd_bus_major = rc;
+
+ nd_class = class_create(THIS_MODULE, "nd");
+ if (IS_ERR(nd_class))
+ goto err_class;
+
+ return 0;
+
+ err_class:
+ unregister_chrdev(nd_bus_major, "ndctl");
+
+ return rc;
+}
+
+void __exit nd_bus_exit(void)
+{
+ class_destroy(nd_class);
+ unregister_chrdev(nd_bus_major, "ndctl");
+}
diff --git a/drivers/block/nd/core.c b/drivers/block/nd/core.c
index 3cccdbc0f3b7..55603ff264ff 100644
--- a/drivers/block/nd/core.c
+++ b/drivers/block/nd/core.c
@@ -13,10 +13,13 @@
#include <linux/export.h>
#include <linux/module.h>
#include <linux/device.h>
+#include <linux/mutex.h>
#include <linux/slab.h>
#include "nd-private.h"
#include "libnd.h"
+LIST_HEAD(nd_bus_list);
+DEFINE_MUTEX(nd_bus_list_mutex);
static DEFINE_IDA(nd_ida);
static void nd_bus_release(struct device *dev)
@@ -27,6 +30,54 @@ static void nd_bus_release(struct device *dev)
kfree(nd_bus);
}
+struct nd_bus *to_nd_bus(struct device *dev)
+{
+ struct nd_bus *nd_bus = container_of(dev, struct nd_bus, dev);
+
+ WARN_ON(nd_bus->dev.release != nd_bus_release);
+ return nd_bus;
+}
+EXPORT_SYMBOL_GPL(to_nd_bus);
+
+struct nd_bus_descriptor *to_nd_desc(struct nd_bus *nd_bus)
+{
+ /* struct nd_bus definition is private to libnd */
+ return nd_bus->nd_desc;
+}
+EXPORT_SYMBOL_GPL(to_nd_desc);
+
+static const char *nd_bus_provider(struct nd_bus *nd_bus)
+{
+ struct nd_bus_descriptor *nd_desc = nd_bus->nd_desc;
+ struct device *parent = nd_bus->dev.parent;
+
+ if (nd_desc->provider_name)
+ return nd_desc->provider_name;
+ else if (parent)
+ return dev_name(parent);
+ else
+ return "unknown";
+}
+
+static ssize_t provider_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_bus *nd_bus = to_nd_bus(dev);
+
+ return sprintf(buf, "%s\n", nd_bus_provider(nd_bus));
+}
+static DEVICE_ATTR_RO(provider);
+
+static struct attribute *nd_bus_attributes[] = {
+ &dev_attr_provider.attr,
+ NULL,
+};
+
+struct attribute_group nd_bus_attribute_group = {
+ .attrs = nd_bus_attributes,
+};
+EXPORT_SYMBOL_GPL(nd_bus_attribute_group);
+
struct nd_bus *nd_bus_register(struct device *parent,
struct nd_bus_descriptor *nd_desc)
{
@@ -35,6 +86,7 @@ struct nd_bus *nd_bus_register(struct device *parent,
if (!nd_bus)
return NULL;
+ INIT_LIST_HEAD(&nd_bus->list);
nd_bus->id = ida_simple_get(&nd_ida, 0, 0, GFP_KERNEL);
if (nd_bus->id < 0) {
kfree(nd_bus);
@@ -43,15 +95,26 @@ struct nd_bus *nd_bus_register(struct device *parent,
nd_bus->nd_desc = nd_desc;
nd_bus->dev.parent = parent;
nd_bus->dev.release = nd_bus_release;
+ nd_bus->dev.groups = nd_desc->attr_groups;
dev_set_name(&nd_bus->dev, "ndbus%d", nd_bus->id);
rc = device_register(&nd_bus->dev);
if (rc) {
dev_dbg(&nd_bus->dev, "device registration failed: %d\n", rc);
- put_device(&nd_bus->dev);
- return NULL;
+ goto err;
}
+ rc = nd_bus_create_ndctl(nd_bus);
+ if (rc)
+ goto err;
+
+ mutex_lock(&nd_bus_list_mutex);
+ list_add_tail(&nd_bus->list, &nd_bus_list);
+ mutex_unlock(&nd_bus_list_mutex);
+
return nd_bus;
+ err:
+ put_device(&nd_bus->dev);
+ return NULL;
}
EXPORT_SYMBOL_GPL(nd_bus_register);
@@ -59,9 +122,29 @@ void nd_bus_unregister(struct nd_bus *nd_bus)
{
if (!nd_bus)
return;
+
+ mutex_lock(&nd_bus_list_mutex);
+ list_del_init(&nd_bus->list);
+ mutex_unlock(&nd_bus_list_mutex);
+
+ nd_bus_destroy_ndctl(nd_bus);
+
device_unregister(&nd_bus->dev);
}
EXPORT_SYMBOL_GPL(nd_bus_unregister);
+static __init int libnd_init(void)
+{
+ return nd_bus_init();
+}
+
+static __exit void libnd_exit(void)
+{
+ WARN_ON(!list_empty(&nd_bus_list));
+ nd_bus_exit();
+}
+
MODULE_LICENSE("GPL v2");
MODULE_AUTHOR("Intel Corporation");
+module_init(libnd_init);
+module_exit(libnd_exit);
diff --git a/drivers/block/nd/libnd.h b/drivers/block/nd/libnd.h
index 163832937e9c..86cf3e0573b0 100644
--- a/drivers/block/nd/libnd.h
+++ b/drivers/block/nd/libnd.h
@@ -14,6 +14,8 @@
*/
#ifndef __LIBND_H__
#define __LIBND_H__
+extern struct attribute_group nd_bus_attribute_group;
+
struct nd_dimm;
struct nd_bus_descriptor;
typedef int (*ndctl_fn)(struct nd_bus_descriptor *nd_desc,
@@ -21,6 +23,7 @@ typedef int (*ndctl_fn)(struct nd_bus_descriptor *nd_desc,
unsigned int buf_len);
struct nd_bus_descriptor {
+ const struct attribute_group **attr_groups;
unsigned long dsm_mask;
char *provider_name;
ndctl_fn ndctl;
@@ -30,4 +33,6 @@ struct nd_bus;
struct nd_bus *nd_bus_register(struct device *parent,
struct nd_bus_descriptor *nfit_desc);
void nd_bus_unregister(struct nd_bus *nd_bus);
+struct nd_bus *to_nd_bus(struct device *dev);
+struct nd_bus_descriptor *to_nd_desc(struct nd_bus *nd_bus);
#endif /* __LIBND_H__ */
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
index 3dbab29fa0f9..960dd2f29cdd 100644
--- a/drivers/block/nd/nd-private.h
+++ b/drivers/block/nd/nd-private.h
@@ -17,7 +17,13 @@
struct nd_bus {
struct nd_bus_descriptor *nd_desc;
+ struct list_head list;
struct device dev;
int id;
};
+
+int __init nd_bus_init(void);
+void __exit nd_bus_exit(void);
+int nd_bus_create_ndctl(struct nd_bus *nd_bus);
+void nd_bus_destroy_ndctl(struct nd_bus *nd_bus);
#endif /* __ND_PRIVATE_H__ */
diff --git a/drivers/block/nd/test/nfit.c b/drivers/block/nd/test/nfit.c
index 8691a903515b..1c79f32376fc 100644
--- a/drivers/block/nd/test/nfit.c
+++ b/drivers/block/nd/test/nfit.c
@@ -833,6 +833,8 @@ static void nfit_test1_setup(struct nfit_test *t)
nfit->checksum = nfit_checksum(nfit_buf, size);
}
+extern const struct attribute_group *nd_acpi_attribute_groups[];
+
static int nfit_test_probe(struct platform_device *pdev)
{
struct nd_bus_descriptor *nd_desc;
@@ -882,6 +884,7 @@ static int nfit_test_probe(struct platform_device *pdev)
acpi_desc->dev = &pdev->dev;
acpi_desc->nfit = nfit_test->nfit_buf;
nd_desc = &acpi_desc->nd_desc;
+ nd_desc->attr_groups = nd_acpi_attribute_groups;
acpi_desc->nd_bus = nd_bus_register(&pdev->dev, nd_desc);
if (!acpi_desc->nd_bus)
return -ENXIO;
Register the memory devices described in the nfit as libnd 'dimm'
devices on an nd bus. The kernel assigned device id for dimms is
dynamic. If userspace needs a more static identifier it should consult
a provider-specific attribute. In the case where NFIT is the provider,
the 'nmemX/nfit/handle' or 'nmemX/nfit/serial' attributes may be used
for this purpose.
Cc: Neil Brown <[email protected]>
Cc: <[email protected]>
Cc: Greg KH <[email protected]>
Cc: Robert Moore <[email protected]>
Cc: Rafael J. Wysocki <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/Makefile | 1
drivers/block/nd/acpi.c | 160 +++++++++++++++++++++++++++++++++++++++++
drivers/block/nd/acpi_nfit.h | 1
drivers/block/nd/bus.c | 14 +++-
drivers/block/nd/core.c | 29 +++++++
drivers/block/nd/dimm_devs.c | 92 ++++++++++++++++++++++++
drivers/block/nd/libnd.h | 11 +++
drivers/block/nd/nd-private.h | 12 +++
8 files changed, 318 insertions(+), 2 deletions(-)
create mode 100644 drivers/block/nd/dimm_devs.c
diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
index 7defe18ed009..35e4c1a7a8ff 100644
--- a/drivers/block/nd/Makefile
+++ b/drivers/block/nd/Makefile
@@ -20,3 +20,4 @@ nd_acpi-y := acpi.o
libnd-y := core.o
libnd-y += bus.o
+libnd-y += dimm_devs.o
diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
index dd8505f766ed..af6684341c9b 100644
--- a/drivers/block/nd/acpi.c
+++ b/drivers/block/nd/acpi.c
@@ -369,6 +369,164 @@ const struct attribute_group *nd_acpi_attribute_groups[] = {
};
EXPORT_SYMBOL_GPL(nd_acpi_attribute_groups);
+static struct acpi_nfit_memdev *to_nfit_memdev(struct device *dev)
+{
+ struct nd_dimm *nd_dimm = to_nd_dimm(dev);
+ struct nfit_mem *nfit_mem = nd_dimm_provider_data(nd_dimm);
+
+ return __to_nfit_memdev(nfit_mem);
+}
+
+static struct acpi_nfit_dcr *to_nfit_dcr(struct device *dev)
+{
+ struct nd_dimm *nd_dimm = to_nd_dimm(dev);
+ struct nfit_mem *nfit_mem = nd_dimm_provider_data(nd_dimm);
+
+ return nfit_mem->dcr;
+}
+
+static ssize_t handle_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct acpi_nfit_memdev *memdev = to_nfit_memdev(dev);
+
+ return sprintf(buf, "%#x\n", memdev->nfit_handle);
+}
+static DEVICE_ATTR_RO(handle);
+
+static ssize_t phys_id_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct acpi_nfit_memdev *memdev = to_nfit_memdev(dev);
+
+ return sprintf(buf, "%#x\n", memdev->phys_id);
+}
+static DEVICE_ATTR_RO(phys_id);
+
+static ssize_t vendor_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct acpi_nfit_dcr *dcr = to_nfit_dcr(dev);
+
+ return sprintf(buf, "%#x\n", dcr->vendor_id);
+}
+static DEVICE_ATTR_RO(vendor);
+
+static ssize_t rev_id_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct acpi_nfit_dcr *dcr = to_nfit_dcr(dev);
+
+ return sprintf(buf, "%#x\n", dcr->revision_id);
+}
+static DEVICE_ATTR_RO(rev_id);
+
+static ssize_t device_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct acpi_nfit_dcr *dcr = to_nfit_dcr(dev);
+
+ return sprintf(buf, "%#x\n", dcr->device_id);
+}
+static DEVICE_ATTR_RO(device);
+
+static ssize_t format_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct acpi_nfit_dcr *dcr = to_nfit_dcr(dev);
+
+ return sprintf(buf, "%#x\n", dcr->fic);
+}
+static DEVICE_ATTR_RO(format);
+
+static ssize_t serial_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct acpi_nfit_dcr *dcr = to_nfit_dcr(dev);
+
+ return sprintf(buf, "%#x\n", dcr->serial_number);
+}
+static DEVICE_ATTR_RO(serial);
+
+static struct attribute *nd_acpi_dimm_attributes[] = {
+ &dev_attr_handle.attr,
+ &dev_attr_phys_id.attr,
+ &dev_attr_vendor.attr,
+ &dev_attr_device.attr,
+ &dev_attr_format.attr,
+ &dev_attr_serial.attr,
+ &dev_attr_rev_id.attr,
+ NULL,
+};
+
+static umode_t nd_acpi_dimm_attr_visible(struct kobject *kobj, struct attribute *a, int n)
+{
+ struct device *dev = container_of(kobj, struct device, kobj);
+
+ if (to_nfit_dcr(dev))
+ return a->mode;
+ else
+ return 0;
+}
+
+static struct attribute_group nd_acpi_dimm_attribute_group = {
+ .name = "nfit",
+ .attrs = nd_acpi_dimm_attributes,
+ .is_visible = nd_acpi_dimm_attr_visible,
+};
+
+static const struct attribute_group *nd_acpi_dimm_attribute_groups[] = {
+ &nd_acpi_dimm_attribute_group,
+ NULL,
+};
+
+static struct nd_dimm *nd_acpi_dimm_by_handle(struct acpi_nfit_desc *acpi_desc,
+ u32 nfit_handle)
+{
+ struct nfit_mem *nfit_mem;
+
+ list_for_each_entry(nfit_mem, &acpi_desc->dimms, list)
+ if (__to_nfit_memdev(nfit_mem)->nfit_handle == nfit_handle)
+ return nfit_mem->nd_dimm;
+
+ return NULL;
+}
+
+static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
+{
+ struct nfit_mem *nfit_mem;
+
+ list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) {
+ struct nd_dimm *nd_dimm;
+ unsigned long flags = 0;
+ u32 nfit_handle;
+
+ nfit_handle = __to_nfit_memdev(nfit_mem)->nfit_handle;
+ nd_dimm = nd_acpi_dimm_by_handle(acpi_desc, nfit_handle);
+ if (nd_dimm) {
+ /*
+ * If for some reason we find multiple DCRs the
+ * first one wins
+ */
+ dev_err(acpi_desc->dev, "duplicate DCR detected: %s\n",
+ nd_dimm_name(nd_dimm));
+ continue;
+ }
+
+ if (nfit_mem->bdw && nfit_mem->memdev_pmem)
+ flags |= NDD_ALIASING;
+
+ nd_dimm = nd_dimm_create(acpi_desc->nd_bus, nfit_mem,
+ nd_acpi_dimm_attribute_groups, flags);
+ if (!nd_dimm)
+ return -ENOMEM;
+
+ nfit_mem->nd_dimm = nd_dimm;
+ }
+
+ return 0;
+}
+
int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
{
struct device *dev = acpi_desc->dev;
@@ -406,7 +564,7 @@ int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
if (nfit_mem_init(acpi_desc) != 0)
return -ENOMEM;
- return 0;
+ return nd_acpi_register_dimms(acpi_desc);
}
EXPORT_SYMBOL_GPL(nd_acpi_nfit_init);
diff --git a/drivers/block/nd/acpi_nfit.h b/drivers/block/nd/acpi_nfit.h
index b65745ca3cbc..00aaaee5953b 100644
--- a/drivers/block/nd/acpi_nfit.h
+++ b/drivers/block/nd/acpi_nfit.h
@@ -233,6 +233,7 @@ struct nfit_memdev {
/* assembled tables for a given dimm/memory-device */
struct nfit_mem {
+ struct nd_dimm *nd_dimm;
struct acpi_nfit_memdev *memdev_dcr;
struct acpi_nfit_memdev *memdev_pmem;
struct acpi_nfit_dcr *dcr;
diff --git a/drivers/block/nd/bus.c b/drivers/block/nd/bus.c
index 635f2e926426..ee56aa1ab2ad 100644
--- a/drivers/block/nd/bus.c
+++ b/drivers/block/nd/bus.c
@@ -13,6 +13,7 @@
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/uaccess.h>
#include <linux/fcntl.h>
+#include <linux/async.h>
#include <linux/slab.h>
#include <linux/fs.h>
#include <linux/io.h>
@@ -21,6 +22,10 @@
static int nd_bus_major;
static struct class *nd_class;
+struct bus_type nd_bus_type = {
+ .name = "nd",
+};
+
int nd_bus_create_ndctl(struct nd_bus *nd_bus)
{
dev_t devt = MKDEV(nd_bus_major, nd_bus->id);
@@ -59,9 +64,13 @@ int __init nd_bus_init(void)
{
int rc;
+ rc = bus_register(&nd_bus_type);
+ if (rc)
+ return rc;
+
rc = register_chrdev(0, "ndctl", &nd_bus_fops);
if (rc < 0)
- return rc;
+ goto err_chrdev;
nd_bus_major = rc;
nd_class = class_create(THIS_MODULE, "nd");
@@ -72,6 +81,8 @@ int __init nd_bus_init(void)
err_class:
unregister_chrdev(nd_bus_major, "ndctl");
+ err_chrdev:
+ bus_unregister(&nd_bus_type);
return rc;
}
@@ -80,4 +91,5 @@ void __exit nd_bus_exit(void)
{
class_destroy(nd_class);
unregister_chrdev(nd_bus_major, "ndctl");
+ bus_unregister(&nd_bus_type);
}
diff --git a/drivers/block/nd/core.c b/drivers/block/nd/core.c
index 55603ff264ff..24046db897c1 100644
--- a/drivers/block/nd/core.c
+++ b/drivers/block/nd/core.c
@@ -46,6 +46,19 @@ struct nd_bus_descriptor *to_nd_desc(struct nd_bus *nd_bus)
}
EXPORT_SYMBOL_GPL(to_nd_desc);
+struct nd_bus *walk_to_nd_bus(struct device *nd_dev)
+{
+ struct device *dev;
+
+ for (dev = nd_dev; dev; dev = dev->parent)
+ if (dev->release == nd_bus_release)
+ break;
+ dev_WARN_ONCE(nd_dev, !dev, "invalid dev, not on nd bus\n");
+ if (dev)
+ return to_nd_bus(dev);
+ return NULL;
+}
+
static const char *nd_bus_provider(struct nd_bus *nd_bus)
{
struct nd_bus_descriptor *nd_desc = nd_bus->nd_desc;
@@ -118,6 +131,21 @@ struct nd_bus *nd_bus_register(struct device *parent,
}
EXPORT_SYMBOL_GPL(nd_bus_register);
+static int child_unregister(struct device *dev, void *data)
+{
+ /*
+ * the singular ndctl class device per bus needs to be
+ * "device_destroy"ed, so skip it here
+ *
+ * i.e. remove classless children
+ */
+ if (dev->class)
+ /* pass */;
+ else
+ device_unregister(dev);
+ return 0;
+}
+
void nd_bus_unregister(struct nd_bus *nd_bus)
{
if (!nd_bus)
@@ -127,6 +155,7 @@ void nd_bus_unregister(struct nd_bus *nd_bus)
list_del_init(&nd_bus->list);
mutex_unlock(&nd_bus_list_mutex);
+ device_for_each_child(&nd_bus->dev, NULL, child_unregister);
nd_bus_destroy_ndctl(nd_bus);
device_unregister(&nd_bus->dev);
diff --git a/drivers/block/nd/dimm_devs.c b/drivers/block/nd/dimm_devs.c
new file mode 100644
index 000000000000..19b081392f2f
--- /dev/null
+++ b/drivers/block/nd/dimm_devs.c
@@ -0,0 +1,92 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+#include <linux/device.h>
+#include <linux/slab.h>
+#include <linux/io.h>
+#include <linux/fs.h>
+#include <linux/mm.h>
+#include "nd-private.h"
+
+static DEFINE_IDA(dimm_ida);
+
+static void nd_dimm_release(struct device *dev)
+{
+ struct nd_dimm *nd_dimm = to_nd_dimm(dev);
+
+ ida_simple_remove(&dimm_ida, nd_dimm->id);
+ kfree(nd_dimm);
+}
+
+static struct device_type nd_dimm_device_type = {
+ .name = "nd_dimm",
+ .release = nd_dimm_release,
+};
+
+static bool is_nd_dimm(struct device *dev)
+{
+ return dev->type == &nd_dimm_device_type;
+}
+
+struct nd_dimm *to_nd_dimm(struct device *dev)
+{
+ struct nd_dimm *nd_dimm = container_of(dev, struct nd_dimm, dev);
+
+ WARN_ON(!is_nd_dimm(dev));
+ return nd_dimm;
+}
+EXPORT_SYMBOL_GPL(to_nd_dimm);
+
+const char *nd_dimm_name(struct nd_dimm *nd_dimm)
+{
+ return dev_name(&nd_dimm->dev);
+}
+EXPORT_SYMBOL_GPL(nd_dimm_name);
+
+void *nd_dimm_provider_data(struct nd_dimm *nd_dimm)
+{
+ return nd_dimm->provider_data;
+}
+EXPORT_SYMBOL_GPL(nd_dimm_provider_data);
+
+struct nd_dimm *nd_dimm_create(struct nd_bus *nd_bus, void *provider_data,
+ const struct attribute_group **groups, unsigned long flags)
+{
+ struct nd_dimm *nd_dimm = kzalloc(sizeof(*nd_dimm), GFP_KERNEL);
+ struct device *dev;
+
+ if (!nd_dimm)
+ return NULL;
+
+ nd_dimm->id = ida_simple_get(&dimm_ida, 0, 0, GFP_KERNEL);
+ if (nd_dimm->id < 0) {
+ kfree(nd_dimm);
+ return NULL;
+ }
+ nd_dimm->provider_data = provider_data;
+ nd_dimm->flags = flags;
+
+ dev = &nd_dimm->dev;
+ dev_set_name(dev, "nmem%d", nd_dimm->id);
+ dev->parent = &nd_bus->dev;
+ dev->type = &nd_dimm_device_type;
+ dev->bus = &nd_bus_type;
+ dev->groups = groups;
+ if (device_register(dev) != 0) {
+ put_device(dev);
+ return NULL;
+ }
+
+ return nd_dimm;
+}
+EXPORT_SYMBOL_GPL(nd_dimm_create);
diff --git a/drivers/block/nd/libnd.h b/drivers/block/nd/libnd.h
index 86cf3e0573b0..4f8f2ebbcf7b 100644
--- a/drivers/block/nd/libnd.h
+++ b/drivers/block/nd/libnd.h
@@ -14,6 +14,12 @@
*/
#ifndef __LIBND_H__
#define __LIBND_H__
+
+enum {
+ /* when a dimm supports both PMEM and BLK access a label is required */
+ NDD_ALIASING = 1 << 0,
+};
+
extern struct attribute_group nd_bus_attribute_group;
struct nd_dimm;
@@ -34,5 +40,10 @@ struct nd_bus *nd_bus_register(struct device *parent,
struct nd_bus_descriptor *nfit_desc);
void nd_bus_unregister(struct nd_bus *nd_bus);
struct nd_bus *to_nd_bus(struct device *dev);
+struct nd_dimm *to_nd_dimm(struct device *dev);
struct nd_bus_descriptor *to_nd_desc(struct nd_bus *nd_bus);
+const char *nd_dimm_name(struct nd_dimm *nd_dimm);
+void *nd_dimm_provider_data(struct nd_dimm *nd_dimm);
+struct nd_dimm *nd_dimm_create(struct nd_bus *nd_bus, void *provider_data,
+ const struct attribute_group **groups, unsigned long flags);
#endif /* __LIBND_H__ */
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
index 960dd2f29cdd..cfb5a7241ed1 100644
--- a/drivers/block/nd/nd-private.h
+++ b/drivers/block/nd/nd-private.h
@@ -15,6 +15,10 @@
#include <linux/device.h>
#include "libnd.h"
+extern struct list_head nd_bus_list;
+extern struct mutex nd_bus_list_mutex;
+extern struct bus_type nd_bus_type;
+
struct nd_bus {
struct nd_bus_descriptor *nd_desc;
struct list_head list;
@@ -22,6 +26,14 @@ struct nd_bus {
int id;
};
+struct nd_dimm {
+ unsigned long flags;
+ void *provider_data;
+ struct device dev;
+ int id;
+};
+
+struct nd_bus *walk_to_nd_bus(struct device *nd_dev);
int __init nd_bus_init(void);
void __exit nd_bus_exit(void);
int nd_bus_create_ndctl(struct nd_bus *nd_bus);
Most configuration of the nd-subsystem is done via nd-sysfs attributes.
However, some nd buses, particularly the ACPI.NFIT bus, define a small
set of messages that can be passed to the platform. For convenience we
derivce the initial nd-ioctl-command formats directly from the NFIT DSM
formats.
ND_CMD_SMART: media health and diagnostics
ND_CMD_GET_CONFIG_SIZE: size of the label space
ND_CMD_GET_CONFIG_DATA: read label space
ND_CMD_SET_CONFIG_DATA: write label space
ND_CMD_VENDOR: vendor-specific command passthrough
ND_CMD_ARS_CAP: report address-range-scrubbing capabilities
ND_CMD_START_ARS: initiate scrubbing
ND_CMD_QUERY_ARS: report on scrubbing state
ND_CMD_SMART_THRESHOLD: configure alarm thresholds for smart events
If a platform later defines different commands than this set it is
straightforward to extend support to those formats.
Most of the commands target a specific dimm. However, the
address-range-scrubbing commands target the bus. The 'commands'
attribute in sysfs of an nd-bus, or an nd-dimm enumerate the supported
commands for that object.
Cc: <[email protected]>
Cc: Robert Moore <[email protected]>
Cc: Rafael J. Wysocki <[email protected]>
Reported-by: Nicholas Moulin <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/Kconfig | 12 ++
drivers/block/nd/acpi.c | 237 ++++++++++++++++++++++++++++++
drivers/block/nd/acpi_nfit.h | 3
drivers/block/nd/bus.c | 324 ++++++++++++++++++++++++++++++++++++++++-
drivers/block/nd/core.c | 16 ++
drivers/block/nd/dimm_devs.c | 38 ++++-
drivers/block/nd/libnd.h | 25 +++
drivers/block/nd/nd-private.h | 3
drivers/block/nd/test/nfit.c | 78 ++++++++++
include/uapi/linux/Kbuild | 1
include/uapi/linux/ndctl.h | 178 +++++++++++++++++++++++
11 files changed, 903 insertions(+), 12 deletions(-)
create mode 100644 include/uapi/linux/ndctl.h
diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
index 09f0135147ca..d2d84451e82c 100644
--- a/drivers/block/nd/Kconfig
+++ b/drivers/block/nd/Kconfig
@@ -37,6 +37,18 @@ config ND_ACPI
addition to storage devices this also enables libnd craft
ACPI._DSM messages for platform/dimm configuration.
+config ND_ACPI_DEBUG
+ bool "ACPI: Extra nd_acpi debugging"
+ depends on ND_ACPI
+ depends on DYNAMIC_DEBUG
+ default n
+ help
+ Enabling this option causes the nd_acpi driver to dump the
+ input and output buffers of _DSM operations on the ACPI0012
+ device and its children. This can be very verbose, so leave
+ it disabled unless you are debugging a hardware / firmware
+ issue.
+
config NFIT_TEST
tristate "NFIT TEST: Manufactured NFIT for interface testing"
depends on DMA_CMA
diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
index af6684341c9b..c46e166695f7 100644
--- a/drivers/block/nd/acpi.c
+++ b/drivers/block/nd/acpi.c
@@ -12,6 +12,7 @@
*/
#include <linux/list_sort.h>
#include <linux/module.h>
+#include <linux/ndctl.h>
#include <linux/list.h>
#include <linux/acpi.h>
#include "acpi_nfit.h"
@@ -25,11 +26,160 @@ enum {
NFIT_ACPI_NOTIFY_TABLE = 0x80,
};
+static u8 nd_acpi_uuids[2][16]; /* initialized at nd_acpi_init */
+
+static u8 *nd_acpi_bus_uuid(void)
+{
+ return nd_acpi_uuids[0];
+}
+
+static u8 *nd_acpi_dimm_uuid(void)
+{
+ return nd_acpi_uuids[1];
+}
+
+static struct acpi_nfit_desc *to_acpi_nfit_desc(struct nd_bus_descriptor *nd_desc)
+{
+ return container_of(nd_desc, struct acpi_nfit_desc, nd_desc);
+}
+
+static struct acpi_device *to_acpi_dev(struct acpi_nfit_desc *acpi_desc)
+{
+ struct nd_bus_descriptor *nd_desc = &acpi_desc->nd_desc;
+
+ /*
+ * If provider == 'ACPI.NFIT' we can assume 'dev' is a struct
+ * acpi_device.
+ */
+ if (!nd_desc->provider_name
+ || strcmp(nd_desc->provider_name, "ACPI.NFIT") != 0)
+ return NULL;
+
+ return to_acpi_device(acpi_desc->dev);
+}
+
static int nd_acpi_ctl(struct nd_bus_descriptor *nd_desc,
struct nd_dimm *nd_dimm, unsigned int cmd, void *buf,
unsigned int buf_len)
{
- return -ENOTTY;
+ struct acpi_nfit_desc *acpi_desc = to_acpi_nfit_desc(nd_desc);
+ const struct nd_cmd_desc const *desc = NULL;
+ union acpi_object in_obj, in_buf, *out_obj;
+ struct device *dev = acpi_desc->dev;
+ const char *cmd_name, *dimm_name;
+ unsigned long dsm_mask;
+ acpi_handle handle;
+ u32 offset;
+ int rc, i;
+ u8 *uuid;
+
+ if (nd_dimm) {
+ struct nfit_mem *nfit_mem = nd_dimm_provider_data(nd_dimm);
+ struct acpi_device *adev = nfit_mem->adev;
+
+ if (!adev)
+ return -ENOTTY;
+ dimm_name = dev_name(&adev->dev);
+ cmd_name = nd_dimm_cmd_name(cmd);
+ dsm_mask = nfit_mem->dsm_mask;
+ desc = nd_cmd_dimm_desc(cmd);
+ uuid = nd_acpi_dimm_uuid();
+ handle = adev->handle;
+ } else {
+ struct acpi_device *adev = to_acpi_dev(acpi_desc);
+
+ cmd_name = nd_bus_cmd_name(cmd);
+ dsm_mask = nd_desc->dsm_mask;
+ desc = nd_cmd_bus_desc(cmd);
+ uuid = nd_acpi_bus_uuid();
+ handle = adev->handle;
+ dimm_name = "bus";
+ }
+
+ if (!desc || (cmd && (desc->out_num + desc->in_num == 0)))
+ return -ENOTTY;
+
+ if (!test_bit(cmd, &dsm_mask))
+ return -ENOTTY;
+
+ in_obj.type = ACPI_TYPE_PACKAGE;
+ in_obj.package.count = 1;
+ in_obj.package.elements = &in_buf;
+ in_buf.type = ACPI_TYPE_BUFFER;
+ in_buf.buffer.pointer = buf;
+ in_buf.buffer.length = 0;
+
+ /* libnd has already validated the input envelope */
+ for (i = 0; i < desc->in_num; i++)
+ in_buf.buffer.length += nd_cmd_in_size(nd_dimm, cmd, desc, i, buf);
+
+ dev_dbg(dev, "%s:%s cmd: %s input length: %d\n", __func__, dimm_name,
+ cmd_name, in_buf.buffer.length);
+ if (IS_ENABLED(CONFIG_ND_ACPI_DEBUG))
+ print_hex_dump_debug(cmd_name, DUMP_PREFIX_OFFSET, 4,
+ 4, in_buf.buffer.pointer, min_t(u32, 128,
+ in_buf.buffer.length), true);
+
+ out_obj = acpi_evaluate_dsm(handle, uuid, 1, cmd, &in_obj);
+ if (!out_obj) {
+ dev_dbg(dev, "%s:%s _DSM failed cmd: %s\n", __func__, dimm_name,
+ cmd_name);
+ return -EINVAL;
+ }
+
+ if (out_obj->package.type != ACPI_TYPE_BUFFER) {
+ dev_dbg(dev, "%s:%s unexpected output object type cmd: %s type: %d\n",
+ __func__, dimm_name, cmd_name, out_obj->type);
+ rc = -EINVAL;
+ goto out;
+ }
+
+ dev_dbg(dev, "%s:%s cmd: %s output length: %d\n", __func__, dimm_name,
+ cmd_name, out_obj->buffer.length);
+ if (IS_ENABLED(CONFIG_ND_ACPI_DEBUG))
+ print_hex_dump_debug(cmd_name, DUMP_PREFIX_OFFSET, 4,
+ 4, out_obj->buffer.pointer, min_t(u32, 128,
+ out_obj->buffer.length), true);
+
+ for (i = 0, offset = 0; i < desc->out_num; i++) {
+ u32 out_size = nd_cmd_out_size(nd_dimm, cmd, desc, i, buf,
+ (u32 *) out_obj->buffer.pointer);
+
+ if (offset + out_size > out_obj->buffer.length) {
+ dev_dbg(dev, "%s:%s output object underflow cmd: %s field: %d\n",
+ __func__, dimm_name, cmd_name, i);
+ break;
+ }
+
+ if (in_buf.buffer.length + offset + out_size > buf_len) {
+ dev_dbg(dev, "%s:%s output overrun cmd: %s field: %d\n",
+ __func__, dimm_name, cmd_name, i);
+ rc = -ENXIO;
+ goto out;
+ }
+ memcpy(buf + in_buf.buffer.length + offset,
+ out_obj->buffer.pointer + offset, out_size);
+ offset += out_size;
+ }
+ if (offset + in_buf.buffer.length < buf_len) {
+ if (i >= 1) {
+ /*
+ * status valid, return the number of bytes left
+ * unfilled in the output buffer
+ */
+ rc = buf_len - offset - in_buf.buffer.length;
+ } else {
+ dev_err(dev, "%s:%s underrun cmd: %s buf_len: %d out_len: %d\n",
+ __func__, dimm_name, cmd_name, buf_len, offset);
+ rc = -ENXIO;
+ }
+ } else
+ rc = 0;
+
+ out:
+ ACPI_FREE(out_obj);
+
+ return rc;
}
static const char *spa_type_name(u16 type)
@@ -476,6 +626,7 @@ static struct attribute_group nd_acpi_dimm_attribute_group = {
};
static const struct attribute_group *nd_acpi_dimm_attribute_groups[] = {
+ &nd_dimm_attribute_group,
&nd_acpi_dimm_attribute_group,
NULL,
};
@@ -492,6 +643,50 @@ static struct nd_dimm *nd_acpi_dimm_by_handle(struct acpi_nfit_desc *acpi_desc,
return NULL;
}
+static int nd_acpi_add_dimm(struct acpi_nfit_desc *acpi_desc,
+ struct nfit_mem *nfit_mem, u32 nfit_handle)
+{
+ struct acpi_device *adev, *adev_dimm;
+ struct device *dev = acpi_desc->dev;
+ u8 *uuid = nd_acpi_dimm_uuid();
+ unsigned long long sta;
+ int i, rc = -ENODEV;
+ acpi_status status;
+
+ nfit_mem->dsm_mask = acpi_desc->dimm_dsm_force_en;
+ adev = to_acpi_dev(acpi_desc);
+ if (!adev)
+ return 0;
+
+ adev_dimm = acpi_find_child_device(adev, nfit_handle, false);
+ nfit_mem->adev = adev_dimm;
+ if (!adev_dimm) {
+ dev_err(dev, "no ACPI.NFIT device with _ADR %#x, disabling...\n",
+ nfit_handle);
+ return -ENODEV;
+ }
+
+ status = acpi_evaluate_integer(adev_dimm->handle, "_STA", NULL, &sta);
+ if (status == AE_NOT_FOUND) {
+ dev_dbg(dev, "%s missing _STA, assuming enabled...\n",
+ dev_name(&adev_dimm->dev));
+ rc = 0;
+ } else if (ACPI_FAILURE(status))
+ dev_err(dev, "%s failed to retrieve_STA, disabling...\n",
+ dev_name(&adev_dimm->dev));
+ else if ((sta & ACPI_STA_DEVICE_ENABLED) == 0)
+ dev_info(dev, "%s disabled by firmware\n",
+ dev_name(&adev_dimm->dev));
+ else
+ rc = 0;
+
+ for (i = ND_CMD_SMART; i <= ND_CMD_VENDOR; i++)
+ if (acpi_check_dsm(adev_dimm->handle, uuid, 1, 1ULL << i))
+ set_bit(i, &nfit_mem->dsm_mask);
+
+ return rc;
+}
+
static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
{
struct nfit_mem *nfit_mem;
@@ -500,6 +695,7 @@ static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
struct nd_dimm *nd_dimm;
unsigned long flags = 0;
u32 nfit_handle;
+ int rc;
nfit_handle = __to_nfit_memdev(nfit_mem)->nfit_handle;
nd_dimm = nd_acpi_dimm_by_handle(acpi_desc, nfit_handle);
@@ -516,8 +712,13 @@ static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
if (nfit_mem->bdw && nfit_mem->memdev_pmem)
flags |= NDD_ALIASING;
+ rc = nd_acpi_add_dimm(acpi_desc, nfit_mem, nfit_handle);
+ if (rc)
+ continue;
+
nd_dimm = nd_dimm_create(acpi_desc->nd_bus, nfit_mem,
- nd_acpi_dimm_attribute_groups, flags);
+ nd_acpi_dimm_attribute_groups,
+ flags, &nfit_mem->dsm_mask);
if (!nd_dimm)
return -ENOMEM;
@@ -527,6 +728,22 @@ static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
return 0;
}
+static void nd_acpi_init_dsms(struct acpi_nfit_desc *acpi_desc)
+{
+ struct nd_bus_descriptor *nd_desc = &acpi_desc->nd_desc;
+ u8 *uuid = nd_acpi_bus_uuid();
+ struct acpi_device *adev;
+ int i;
+
+ adev = to_acpi_dev(acpi_desc);
+ if (!adev)
+ return;
+
+ for (i = ND_CMD_ARS_CAP; i <= ND_CMD_ARS_QUERY; i++)
+ if (acpi_check_dsm(adev->handle, uuid, 1, 1ULL << i))
+ set_bit(i, &nd_desc->dsm_mask);
+}
+
int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
{
struct device *dev = acpi_desc->dev;
@@ -564,6 +781,8 @@ int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
if (nfit_mem_init(acpi_desc) != 0)
return -ENOMEM;
+ nd_acpi_init_dsms(acpi_desc);
+
return nd_acpi_register_dimms(acpi_desc);
}
EXPORT_SYMBOL_GPL(nd_acpi_nfit_init);
@@ -641,6 +860,14 @@ static struct acpi_driver nd_acpi_driver = {
static __init int nd_acpi_init(void)
{
+ char *uuids[] = {
+ /* bus interface */
+ "2f10e7a4-9e91-11e4-89d3-123b93f75cba",
+ /* per-dimm interface */
+ "4309ac30-0d11-11e4-9191-0800200c9a66",
+ };
+ int i;
+
BUILD_BUG_ON(sizeof(struct acpi_nfit) != 40);
BUILD_BUG_ON(sizeof(struct acpi_nfit_spa) != 56);
BUILD_BUG_ON(sizeof(struct acpi_nfit_memdev) != 48);
@@ -649,6 +876,12 @@ static __init int nd_acpi_init(void)
BUILD_BUG_ON(sizeof(struct acpi_nfit_dcr) != 80);
BUILD_BUG_ON(sizeof(struct acpi_nfit_bdw) != 40);
+ for (i = 0; i < ARRAY_SIZE(uuids); i++)
+ if (acpi_str_to_uuid(uuids[i], nd_acpi_uuids[i]) != AE_OK) {
+ WARN_ON_ONCE(1);
+ return -ENXIO;
+ }
+
return acpi_bus_register_driver(&nd_acpi_driver);
}
diff --git a/drivers/block/nd/acpi_nfit.h b/drivers/block/nd/acpi_nfit.h
index 00aaaee5953b..2faac336c07d 100644
--- a/drivers/block/nd/acpi_nfit.h
+++ b/drivers/block/nd/acpi_nfit.h
@@ -241,6 +241,8 @@ struct nfit_mem {
struct acpi_nfit_spa *spa_dcr;
struct acpi_nfit_spa *spa_bdw;
struct list_head list;
+ struct acpi_device *adev;
+ unsigned long dsm_mask;
};
struct acpi_nfit_desc {
@@ -253,6 +255,7 @@ struct acpi_nfit_desc {
struct list_head bdws;
struct nd_bus *nd_bus;
struct device *dev;
+ unsigned long dimm_dsm_force_en;
};
static inline struct acpi_nfit_memdev *__to_nfit_memdev(struct nfit_mem *nfit_mem)
diff --git a/drivers/block/nd/bus.c b/drivers/block/nd/bus.c
index ee56aa1ab2ad..a271e01af4a9 100644
--- a/drivers/block/nd/bus.c
+++ b/drivers/block/nd/bus.c
@@ -11,14 +11,18 @@
* General Public License for more details.
*/
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+#include <linux/vmalloc.h>
#include <linux/uaccess.h>
#include <linux/fcntl.h>
#include <linux/async.h>
+#include <linux/ndctl.h>
#include <linux/slab.h>
#include <linux/fs.h>
#include <linux/io.h>
+#include <linux/mm.h>
#include "nd-private.h"
+int nd_dimm_major;
static int nd_bus_major;
static struct class *nd_class;
@@ -47,19 +51,323 @@ void nd_bus_destroy_ndctl(struct nd_bus *nd_bus)
device_destroy(nd_class, MKDEV(nd_bus_major, nd_bus->id));
}
+static const struct nd_cmd_desc const __nd_cmd_dimm_descs[] = {
+ [ND_CMD_IMPLEMENTED] = { },
+ [ND_CMD_SMART] = {
+ .out_num = 2,
+ .out_sizes = { 4, 8, },
+ },
+ [ND_CMD_SMART_THRESHOLD] = {
+ .out_num = 2,
+ .out_sizes = { 4, 8, },
+ },
+ [ND_CMD_DIMM_FLAGS] = {
+ .out_num = 2,
+ .out_sizes = { 4, 4 },
+ },
+ [ND_CMD_GET_CONFIG_SIZE] = {
+ .out_num = 3,
+ .out_sizes = { 4, 4, 4, },
+ },
+ [ND_CMD_GET_CONFIG_DATA] = {
+ .in_num = 2,
+ .in_sizes = { 4, 4, },
+ .out_num = 2,
+ .out_sizes = { 4, UINT_MAX, },
+ },
+ [ND_CMD_SET_CONFIG_DATA] = {
+ .in_num = 3,
+ .in_sizes = { 4, 4, UINT_MAX, },
+ .out_num = 1,
+ .out_sizes = { 4, },
+ },
+ [ND_CMD_VENDOR] = {
+ .in_num = 3,
+ .in_sizes = { 4, 4, UINT_MAX, },
+ .out_num = 3,
+ .out_sizes = { 4, 4, UINT_MAX, },
+ },
+};
+
+struct nd_cmd_desc const *nd_cmd_dimm_desc(int cmd)
+{
+ if (cmd < ARRAY_SIZE(__nd_cmd_dimm_descs))
+ return &__nd_cmd_dimm_descs[cmd];
+ return NULL;
+}
+EXPORT_SYMBOL_GPL(nd_cmd_dimm_desc);
+
+static const struct nd_cmd_desc const __nd_cmd_bus_descs[] = {
+ [ND_CMD_IMPLEMENTED] = { },
+ [ND_CMD_ARS_CAP] = {
+ .in_num = 2,
+ .in_sizes = { 8, 8, },
+ .out_num = 2,
+ .out_sizes = { 4, 4, },
+ },
+ [ND_CMD_ARS_START] = {
+ .in_num = 4,
+ .in_sizes = { 8, 8, 2, 6, },
+ .out_num = 1,
+ .out_sizes = { 4, },
+ },
+ [ND_CMD_ARS_QUERY] = {
+ .out_num = 2,
+ .out_sizes = { 4, UINT_MAX, },
+ },
+};
+
+const struct nd_cmd_desc *nd_cmd_bus_desc(int cmd)
+{
+ if (cmd < ARRAY_SIZE(__nd_cmd_bus_descs))
+ return &__nd_cmd_bus_descs[cmd];
+ return NULL;
+}
+EXPORT_SYMBOL_GPL(nd_cmd_bus_desc);
+
+u32 nd_cmd_in_size(struct nd_dimm *nd_dimm, int cmd,
+ const struct nd_cmd_desc *desc, int idx, void *buf)
+{
+ if (idx >= desc->in_num)
+ return UINT_MAX;
+
+ if (desc->in_sizes[idx] < UINT_MAX)
+ return desc->in_sizes[idx];
+
+ if (nd_dimm && cmd == ND_CMD_SET_CONFIG_DATA && idx == 2) {
+ struct nd_cmd_set_config_hdr *hdr = buf;
+
+ return hdr->in_length;
+ } else if (nd_dimm && cmd == ND_CMD_VENDOR && idx == 2) {
+ struct nd_cmd_vendor_hdr *hdr = buf;
+
+ return hdr->in_length;
+ }
+
+ return UINT_MAX;
+}
+EXPORT_SYMBOL_GPL(nd_cmd_in_size);
+
+u32 nd_cmd_out_size(struct nd_dimm *nd_dimm, int cmd,
+ const struct nd_cmd_desc *desc, int idx, const u32 *in_field,
+ const u32 *out_field)
+{
+ if (idx >= desc->out_num)
+ return UINT_MAX;
+
+ if (desc->out_sizes[idx] < UINT_MAX)
+ return desc->out_sizes[idx];
+
+ if (nd_dimm && cmd == ND_CMD_GET_CONFIG_DATA && idx == 1)
+ return in_field[1];
+ else if (nd_dimm && cmd == ND_CMD_VENDOR && idx == 2)
+ return out_field[1];
+ else if (!nd_dimm && cmd == ND_CMD_ARS_QUERY && idx == 1)
+ return ND_CMD_ARS_QUERY_MAX;
+
+ return UINT_MAX;
+}
+EXPORT_SYMBOL_GPL(nd_cmd_out_size);
+
+static int __nd_ioctl(struct nd_bus *nd_bus, struct nd_dimm *nd_dimm,
+ int read_only, unsigned int ioctl_cmd, unsigned long arg)
+{
+ struct nd_bus_descriptor *nd_desc = nd_bus->nd_desc;
+ size_t buf_len = 0, in_len = 0, out_len = 0;
+ static char out_env[ND_CMD_MAX_ENVELOPE];
+ static char in_env[ND_CMD_MAX_ENVELOPE];
+ const struct nd_cmd_desc *desc = NULL;
+ unsigned int cmd = _IOC_NR(ioctl_cmd);
+ void __user *p = (void __user *) arg;
+ struct device *dev = &nd_bus->dev;
+ const char *cmd_name, *dimm_name;
+ unsigned long dsm_mask;
+ void *buf;
+ int rc, i;
+
+ if (nd_dimm) {
+ desc = nd_cmd_dimm_desc(cmd);
+ cmd_name = nd_dimm_cmd_name(cmd);
+ dsm_mask = nd_dimm->dsm_mask ? *(nd_dimm->dsm_mask) : 0;
+ dimm_name = dev_name(&nd_dimm->dev);
+ } else {
+ desc = nd_cmd_bus_desc(cmd);
+ cmd_name = nd_bus_cmd_name(cmd);
+ dsm_mask = nd_desc->dsm_mask;
+ dimm_name = "bus";
+ }
+
+ if (!desc || (desc->out_num + desc->in_num == 0) ||
+ !test_bit(cmd, &dsm_mask))
+ return -ENOTTY;
+
+ /* fail write commands (when read-only) */
+ if (read_only)
+ switch (ioctl_cmd) {
+ case ND_IOCTL_VENDOR:
+ case ND_IOCTL_SET_CONFIG_DATA:
+ case ND_IOCTL_ARS_START:
+ dev_dbg(&nd_bus->dev, "'%s' command while read-only.\n",
+ nd_dimm ? nd_dimm_cmd_name(cmd)
+ : nd_bus_cmd_name(cmd));
+ return -EPERM;
+ default:
+ break;
+ }
+
+ /* process an input envelope */
+ for (i = 0; i < desc->in_num; i++) {
+ u32 in_size, copy;
+
+ in_size = nd_cmd_in_size(nd_dimm, cmd, desc, i, in_env);
+ if (in_size == UINT_MAX) {
+ dev_err(dev, "%s:%s unknown input size cmd: %s field: %d\n",
+ __func__, dimm_name, cmd_name, i);
+ return -ENXIO;
+ }
+ if (!access_ok(VERIFY_READ, p + in_len, in_size))
+ return -EFAULT;
+ if (in_len < sizeof(in_env))
+ copy = min_t(u32, sizeof(in_env) - in_len, in_size);
+ else
+ copy = 0;
+ if (copy && copy_from_user(&in_env[in_len], p + in_len, copy))
+ return -EFAULT;
+ in_len += in_size;
+ }
+
+ /* process an output envelope */
+ for (i = 0; i < desc->out_num; i++) {
+ u32 out_size = nd_cmd_out_size(nd_dimm, cmd, desc, i,
+ (u32 *) in_env, (u32 *) out_env);
+ u32 copy;
+
+ if (out_size == UINT_MAX) {
+ dev_dbg(dev, "%s:%s unknown output size cmd: %s field: %d\n",
+ __func__, dimm_name, cmd_name, i);
+ return -EFAULT;
+ }
+ if (!access_ok(VERIFY_WRITE, p + in_len + out_len, out_size))
+ return -EFAULT;
+ if (out_len < sizeof(out_env))
+ copy = min_t(u32, sizeof(out_env) - out_len, out_size);
+ else
+ copy = 0;
+ if (copy && copy_from_user(&out_env[out_len], p + in_len + out_len,
+ copy))
+ return -EFAULT;
+ out_len += out_size;
+ }
+
+ buf_len = out_len + in_len;
+ if (!access_ok(VERIFY_WRITE, p, sizeof(buf_len)))
+ return -EFAULT;
+
+ if (buf_len > ND_IOCTL_MAX_BUFLEN) {
+ dev_dbg(dev, "%s:%s cmd: %s buf_len: %zd > %d\n", __func__,
+ dimm_name, cmd_name, buf_len,
+ ND_IOCTL_MAX_BUFLEN);
+ return -EINVAL;
+ }
+
+ buf = vmalloc(buf_len);
+ if (!buf)
+ return -ENOMEM;
+
+ if (copy_from_user(buf, p, buf_len)) {
+ rc = -EFAULT;
+ goto out;
+ }
+
+ rc = nd_desc->ndctl(nd_desc, nd_dimm, cmd, buf, buf_len);
+ if (rc < 0)
+ goto out;
+ if (copy_to_user(p, buf, buf_len))
+ rc = -EFAULT;
+ out:
+ vfree(buf);
+ return rc;
+}
+
static long nd_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
- return -ENXIO;
+ long id = (long) file->private_data;
+ int rc = -ENXIO, read_only;
+ struct nd_bus *nd_bus;
+
+ read_only = (O_RDWR != (file->f_flags & O_ACCMODE));
+ mutex_lock(&nd_bus_list_mutex);
+ list_for_each_entry(nd_bus, &nd_bus_list, list) {
+ if (nd_bus->id == id) {
+ rc = __nd_ioctl(nd_bus, NULL, read_only, cmd, arg);
+ break;
+ }
+ }
+ mutex_unlock(&nd_bus_list_mutex);
+
+ return rc;
+}
+
+static int match_dimm(struct device *dev, void *data)
+{
+ long id = (long) data;
+
+ if (is_nd_dimm(dev)) {
+ struct nd_dimm *nd_dimm = to_nd_dimm(dev);
+
+ return nd_dimm->id == id;
+ }
+
+ return 0;
+}
+
+static long nd_dimm_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+ int rc = -ENXIO, read_only;
+ struct nd_bus *nd_bus;
+
+ read_only = (O_RDWR != (file->f_flags & O_ACCMODE));
+ mutex_lock(&nd_bus_list_mutex);
+ list_for_each_entry(nd_bus, &nd_bus_list, list) {
+ struct device *dev = device_find_child(&nd_bus->dev,
+ file->private_data, match_dimm);
+
+ if (!dev)
+ continue;
+
+ rc = __nd_ioctl(nd_bus, to_nd_dimm(dev), read_only, cmd, arg);
+ put_device(dev);
+ break;
+ }
+ mutex_unlock(&nd_bus_list_mutex);
+
+ return rc;
+}
+
+static int nd_open(struct inode *inode, struct file *file)
+{
+ long minor = iminor(inode);
+
+ file->private_data = (void *) minor;
+ return 0;
}
static const struct file_operations nd_bus_fops = {
.owner = THIS_MODULE,
- .open = nonseekable_open,
+ .open = nd_open,
.unlocked_ioctl = nd_ioctl,
.compat_ioctl = nd_ioctl,
.llseek = noop_llseek,
};
+static const struct file_operations nd_dimm_fops = {
+ .owner = THIS_MODULE,
+ .open = nd_open,
+ .unlocked_ioctl = nd_dimm_ioctl,
+ .compat_ioctl = nd_dimm_ioctl,
+ .llseek = noop_llseek,
+};
+
int __init nd_bus_init(void)
{
int rc;
@@ -70,9 +378,14 @@ int __init nd_bus_init(void)
rc = register_chrdev(0, "ndctl", &nd_bus_fops);
if (rc < 0)
- goto err_chrdev;
+ goto err_bus_chrdev;
nd_bus_major = rc;
+ rc = register_chrdev(0, "dimmctl", &nd_dimm_fops);
+ if (rc < 0)
+ goto err_dimm_chrdev;
+ nd_dimm_major = rc;
+
nd_class = class_create(THIS_MODULE, "nd");
if (IS_ERR(nd_class))
goto err_class;
@@ -80,8 +393,10 @@ int __init nd_bus_init(void)
return 0;
err_class:
+ unregister_chrdev(nd_dimm_major, "dimmctl");
+ err_dimm_chrdev:
unregister_chrdev(nd_bus_major, "ndctl");
- err_chrdev:
+ err_bus_chrdev:
bus_unregister(&nd_bus_type);
return rc;
@@ -91,5 +406,6 @@ void __exit nd_bus_exit(void)
{
class_destroy(nd_class);
unregister_chrdev(nd_bus_major, "ndctl");
+ unregister_chrdev(nd_dimm_major, "dimmctl");
bus_unregister(&nd_bus_type);
}
diff --git a/drivers/block/nd/core.c b/drivers/block/nd/core.c
index 24046db897c1..b14bf47dc292 100644
--- a/drivers/block/nd/core.c
+++ b/drivers/block/nd/core.c
@@ -13,6 +13,7 @@
#include <linux/export.h>
#include <linux/module.h>
#include <linux/device.h>
+#include <linux/ndctl.h>
#include <linux/mutex.h>
#include <linux/slab.h>
#include "nd-private.h"
@@ -59,6 +60,20 @@ struct nd_bus *walk_to_nd_bus(struct device *nd_dev)
return NULL;
}
+static ssize_t commands_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ int cmd, len = 0;
+ struct nd_bus *nd_bus = to_nd_bus(dev);
+ struct nd_bus_descriptor *nd_desc = nd_bus->nd_desc;
+
+ for_each_set_bit(cmd, &nd_desc->dsm_mask, BITS_PER_LONG)
+ len += sprintf(buf + len, "%s ", nd_bus_cmd_name(cmd));
+ len += sprintf(buf + len, "\n");
+ return len;
+}
+static DEVICE_ATTR_RO(commands);
+
static const char *nd_bus_provider(struct nd_bus *nd_bus)
{
struct nd_bus_descriptor *nd_desc = nd_bus->nd_desc;
@@ -82,6 +97,7 @@ static ssize_t provider_show(struct device *dev,
static DEVICE_ATTR_RO(provider);
static struct attribute *nd_bus_attributes[] = {
+ &dev_attr_commands.attr,
&dev_attr_provider.attr,
NULL,
};
diff --git a/drivers/block/nd/dimm_devs.c b/drivers/block/nd/dimm_devs.c
index 19b081392f2f..3fa26f61c3db 100644
--- a/drivers/block/nd/dimm_devs.c
+++ b/drivers/block/nd/dimm_devs.c
@@ -12,6 +12,7 @@
*/
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/device.h>
+#include <linux/ndctl.h>
#include <linux/slab.h>
#include <linux/io.h>
#include <linux/fs.h>
@@ -33,7 +34,7 @@ static struct device_type nd_dimm_device_type = {
.release = nd_dimm_release,
};
-static bool is_nd_dimm(struct device *dev)
+bool is_nd_dimm(struct device *dev)
{
return dev->type == &nd_dimm_device_type;
}
@@ -55,12 +56,41 @@ EXPORT_SYMBOL_GPL(nd_dimm_name);
void *nd_dimm_provider_data(struct nd_dimm *nd_dimm)
{
- return nd_dimm->provider_data;
+ if (nd_dimm)
+ return nd_dimm->provider_data;
+ return NULL;
}
EXPORT_SYMBOL_GPL(nd_dimm_provider_data);
+static ssize_t commands_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_dimm *nd_dimm = to_nd_dimm(dev);
+ int cmd, len = 0;
+
+ if (!nd_dimm->dsm_mask)
+ return sprintf(buf, "\n");
+
+ for_each_set_bit(cmd, nd_dimm->dsm_mask, BITS_PER_LONG)
+ len += sprintf(buf + len, "%s ", nd_dimm_cmd_name(cmd));
+ len += sprintf(buf + len, "\n");
+ return len;
+}
+static DEVICE_ATTR_RO(commands);
+
+static struct attribute *nd_dimm_attributes[] = {
+ &dev_attr_commands.attr,
+ NULL,
+};
+
+struct attribute_group nd_dimm_attribute_group = {
+ .attrs = nd_dimm_attributes,
+};
+EXPORT_SYMBOL_GPL(nd_dimm_attribute_group);
+
struct nd_dimm *nd_dimm_create(struct nd_bus *nd_bus, void *provider_data,
- const struct attribute_group **groups, unsigned long flags)
+ const struct attribute_group **groups, unsigned long flags,
+ unsigned long *dsm_mask)
{
struct nd_dimm *nd_dimm = kzalloc(sizeof(*nd_dimm), GFP_KERNEL);
struct device *dev;
@@ -75,12 +105,14 @@ struct nd_dimm *nd_dimm_create(struct nd_bus *nd_bus, void *provider_data,
}
nd_dimm->provider_data = provider_data;
nd_dimm->flags = flags;
+ nd_dimm->dsm_mask = dsm_mask;
dev = &nd_dimm->dev;
dev_set_name(dev, "nmem%d", nd_dimm->id);
dev->parent = &nd_bus->dev;
dev->type = &nd_dimm_device_type;
dev->bus = &nd_bus_type;
+ dev->devt = MKDEV(nd_dimm_major, nd_dimm->id);
dev->groups = groups;
if (device_register(dev) != 0) {
put_device(dev);
diff --git a/drivers/block/nd/libnd.h b/drivers/block/nd/libnd.h
index 4f8f2ebbcf7b..23dae8c1e65b 100644
--- a/drivers/block/nd/libnd.h
+++ b/drivers/block/nd/libnd.h
@@ -14,13 +14,21 @@
*/
#ifndef __LIBND_H__
#define __LIBND_H__
+#include <linux/sizes.h>
enum {
/* when a dimm supports both PMEM and BLK access a label is required */
NDD_ALIASING = 1 << 0,
+
+ /* need to set a limit somewhere, but yes, this is likely overkill */
+ ND_IOCTL_MAX_BUFLEN = SZ_4M,
+ ND_CMD_MAX_ELEM = 4,
+ ND_CMD_MAX_ENVELOPE = 16,
+ ND_CMD_ARS_QUERY_MAX = SZ_4K,
};
extern struct attribute_group nd_bus_attribute_group;
+extern struct attribute_group nd_dimm_attribute_group;
struct nd_dimm;
struct nd_bus_descriptor;
@@ -35,6 +43,13 @@ struct nd_bus_descriptor {
ndctl_fn ndctl;
};
+struct nd_cmd_desc {
+ int in_num;
+ int out_num;
+ u32 in_sizes[ND_CMD_MAX_ELEM];
+ int out_sizes[ND_CMD_MAX_ELEM];
+};
+
struct nd_bus;
struct nd_bus *nd_bus_register(struct device *parent,
struct nd_bus_descriptor *nfit_desc);
@@ -45,5 +60,13 @@ struct nd_bus_descriptor *to_nd_desc(struct nd_bus *nd_bus);
const char *nd_dimm_name(struct nd_dimm *nd_dimm);
void *nd_dimm_provider_data(struct nd_dimm *nd_dimm);
struct nd_dimm *nd_dimm_create(struct nd_bus *nd_bus, void *provider_data,
- const struct attribute_group **groups, unsigned long flags);
+ const struct attribute_group **groups, unsigned long flags,
+ unsigned long *dsm_mask);
+const struct nd_cmd_desc const *nd_cmd_dimm_desc(int cmd);
+const struct nd_cmd_desc const *nd_cmd_bus_desc(int cmd);
+u32 nd_cmd_in_size(struct nd_dimm *nd_dimm, int cmd,
+ const struct nd_cmd_desc *desc, int idx, void *buf);
+u32 nd_cmd_out_size(struct nd_dimm *nd_dimm, int cmd,
+ const struct nd_cmd_desc *desc, int idx, const u32 *in_field,
+ const u32 *out_field);
#endif /* __LIBND_H__ */
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
index cfb5a7241ed1..968e5089f72c 100644
--- a/drivers/block/nd/nd-private.h
+++ b/drivers/block/nd/nd-private.h
@@ -18,6 +18,7 @@
extern struct list_head nd_bus_list;
extern struct mutex nd_bus_list_mutex;
extern struct bus_type nd_bus_type;
+extern int nd_dimm_major;
struct nd_bus {
struct nd_bus_descriptor *nd_desc;
@@ -29,10 +30,12 @@ struct nd_bus {
struct nd_dimm {
unsigned long flags;
void *provider_data;
+ unsigned long *dsm_mask;
struct device dev;
int id;
};
+bool is_nd_dimm(struct device *dev);
struct nd_bus *walk_to_nd_bus(struct device *nd_dev);
int __init nd_bus_init(void);
void __exit nd_bus_exit(void);
diff --git a/drivers/block/nd/test/nfit.c b/drivers/block/nd/test/nfit.c
index 1c79f32376fc..50916f0ca901 100644
--- a/drivers/block/nd/test/nfit.c
+++ b/drivers/block/nd/test/nfit.c
@@ -14,6 +14,7 @@
#include <linux/platform_device.h>
#include <linux/dma-mapping.h>
#include <linux/module.h>
+#include <linux/ndctl.h>
#include <linux/sizes.h>
#include <linux/slab.h>
#include "nfit_test.h"
@@ -142,7 +143,74 @@ static int nfit_test_ctl(struct nd_bus_descriptor *nd_desc,
struct nd_dimm *nd_dimm, unsigned int cmd, void *buf,
unsigned int buf_len)
{
- return -ENOTTY;
+ struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc);
+ struct nfit_test *t = container_of(acpi_desc, typeof(*t), acpi_desc);
+ struct nfit_mem *nfit_mem = nd_dimm_provider_data(nd_dimm);
+ int i, rc;
+
+ if (!nfit_mem || !test_bit(cmd, &nfit_mem->dsm_mask))
+ return -ENXIO;
+
+ /* lookup label space for the given dimm */
+ for (i = 0; i < ARRAY_SIZE(handle); i++)
+ if (__to_nfit_memdev(nfit_mem)->nfit_handle == handle[i])
+ break;
+ if (i >= ARRAY_SIZE(handle))
+ return -ENXIO;
+
+ switch (cmd) {
+ case ND_CMD_GET_CONFIG_SIZE: {
+ struct nd_cmd_get_config_size *nd_cmd = buf;
+
+ if (buf_len < sizeof(*nd_cmd))
+ return -EINVAL;
+ nd_cmd->status = 0;
+ nd_cmd->config_size = LABEL_SIZE;
+ nd_cmd->max_xfer = SZ_4K;
+ rc = 0;
+ break;
+ }
+ case ND_CMD_GET_CONFIG_DATA: {
+ struct nd_cmd_get_config_data_hdr *nd_cmd = buf;
+ unsigned int len, offset = nd_cmd->in_offset;
+
+ if (buf_len < sizeof(*nd_cmd))
+ return -EINVAL;
+ if (offset >= LABEL_SIZE)
+ return -EINVAL;
+ if (nd_cmd->in_length + sizeof(*nd_cmd) > buf_len)
+ return -EINVAL;
+
+ nd_cmd->status = 0;
+ len = min(nd_cmd->in_length, LABEL_SIZE - offset);
+ memcpy(nd_cmd->out_buf, t->label[i] + offset, len);
+ rc = buf_len - sizeof(*nd_cmd) - len;
+ break;
+ }
+ case ND_CMD_SET_CONFIG_DATA: {
+ struct nd_cmd_set_config_hdr *nd_cmd = buf;
+ unsigned int len, offset = nd_cmd->in_offset;
+ u32 *status;
+
+ if (buf_len < sizeof(*nd_cmd))
+ return -EINVAL;
+ if (offset >= LABEL_SIZE)
+ return -EINVAL;
+ if (nd_cmd->in_length + sizeof(*nd_cmd) + 4 > buf_len)
+ return -EINVAL;
+
+ status = buf + nd_cmd->in_length + sizeof(*nd_cmd);
+ *status = 0;
+ len = min(nd_cmd->in_length, LABEL_SIZE - offset);
+ memcpy(t->label[i] + offset, nd_cmd->in_buf, len);
+ rc = buf_len - sizeof(*nd_cmd) - (len + 4);
+ break;
+ }
+ default:
+ return -ENOTTY;
+ }
+
+ return rc;
}
static DEFINE_SPINLOCK(nfit_test_lock);
@@ -280,6 +348,7 @@ static int nfit_test0_alloc(struct nfit_test *t)
t->label[i] = test_alloc(t, LABEL_SIZE, &t->label_dma[i]);
if (!t->label[i])
return -ENOMEM;
+ sprintf(t->label[i], "label%d", i);
}
for (i = 0; i < NUM_DCR; i++) {
@@ -322,6 +391,7 @@ static int nfit_test1_alloc(struct nfit_test *t)
static void nfit_test0_setup(struct nfit_test *t)
{
struct nd_bus_descriptor *nd_desc;
+ struct acpi_nfit_desc *acpi_desc;
struct acpi_nfit_memdev *memdev;
void *nfit_buf = t->nfit_buf;
size_t size = t->nfit_size;
@@ -763,7 +833,11 @@ static void nfit_test0_setup(struct nfit_test *t)
nfit->checksum = nfit_checksum(nfit_buf, size);
- nd_desc = &t->acpi_desc.nd_desc;
+ acpi_desc = &t->acpi_desc;
+ set_bit(ND_CMD_GET_CONFIG_SIZE, &acpi_desc->dimm_dsm_force_en);
+ set_bit(ND_CMD_GET_CONFIG_DATA, &acpi_desc->dimm_dsm_force_en);
+ set_bit(ND_CMD_SET_CONFIG_DATA, &acpi_desc->dimm_dsm_force_en);
+ nd_desc = &acpi_desc->nd_desc;
nd_desc->ndctl = nfit_test_ctl;
}
diff --git a/include/uapi/linux/Kbuild b/include/uapi/linux/Kbuild
index 68ceb97c458c..384e8d212b04 100644
--- a/include/uapi/linux/Kbuild
+++ b/include/uapi/linux/Kbuild
@@ -270,6 +270,7 @@ header-y += ncp_fs.h
header-y += ncp.h
header-y += ncp_mount.h
header-y += ncp_no.h
+header-y += ndctl.h
header-y += neighbour.h
header-y += netconf.h
header-y += netdevice.h
diff --git a/include/uapi/linux/ndctl.h b/include/uapi/linux/ndctl.h
new file mode 100644
index 000000000000..62c01bf76198
--- /dev/null
+++ b/include/uapi/linux/ndctl.h
@@ -0,0 +1,178 @@
+/*
+ * Copyright (c) 2014-2015, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU Lesser General Public License,
+ * version 2.1, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT ANY
+ * WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+ * FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for
+ * more details.
+ */
+#ifndef __NDCTL_H__
+#define __NDCTL_H__
+
+#include <linux/types.h>
+
+struct nd_cmd_smart {
+ __u32 status;
+ __u8 data[128];
+} __packed;
+
+struct nd_cmd_smart_threshold {
+ __u32 status;
+ __u8 data[8];
+} __packed;
+
+struct nd_cmd_dimm_flags {
+ __u32 status;
+ __u32 flags;
+} __packed;
+
+struct nd_cmd_get_config_size {
+ __u32 status;
+ __u32 config_size;
+ __u32 max_xfer;
+} __packed;
+
+struct nd_cmd_get_config_data_hdr {
+ __u32 in_offset;
+ __u32 in_length;
+ __u32 status;
+ __u8 out_buf[0];
+} __packed;
+
+struct nd_cmd_set_config_hdr {
+ __u32 in_offset;
+ __u32 in_length;
+ __u8 in_buf[0];
+} __packed;
+
+struct nd_cmd_vendor_hdr {
+ __u32 opcode;
+ __u32 in_length;
+ __u8 in_buf[0];
+} __packed;
+
+struct nd_cmd_vendor_tail {
+ __u32 status;
+ __u32 out_length;
+ __u8 out_buf[0];
+} __packed;
+
+struct nd_cmd_ars_cap {
+ __u64 address;
+ __u64 length;
+ __u32 status;
+ __u32 max_ars_out;
+} __packed;
+
+struct nd_cmd_ars_start {
+ __u64 address;
+ __u64 length;
+ __u16 type;
+ __u8 reserved[6];
+ __u32 status;
+} __packed;
+
+struct nd_cmd_ars_query {
+ __u32 status;
+ __u32 out_length;
+ __u64 address;
+ __u64 length;
+ __u16 type;
+ __u32 num_records;
+ struct nd_ars_record {
+ __u32 handle;
+ __u32 flags;
+ __u64 err_address;
+ __u64 mask;
+ } __packed records[0];
+} __packed;
+
+enum {
+ ND_CMD_IMPLEMENTED = 0,
+
+ /* bus commands */
+ ND_CMD_ARS_CAP = 1,
+ ND_CMD_ARS_START = 2,
+ ND_CMD_ARS_QUERY = 3,
+
+ /* per-dimm commands */
+ ND_CMD_SMART = 1,
+ ND_CMD_SMART_THRESHOLD = 2,
+ ND_CMD_DIMM_FLAGS = 3,
+ ND_CMD_GET_CONFIG_SIZE = 4,
+ ND_CMD_GET_CONFIG_DATA = 5,
+ ND_CMD_SET_CONFIG_DATA = 6,
+ ND_CMD_VENDOR_EFFECT_LOG_SIZE = 7,
+ ND_CMD_VENDOR_EFFECT_LOG = 8,
+ ND_CMD_VENDOR = 9,
+};
+
+static inline const char *nd_bus_cmd_name(unsigned cmd)
+{
+ static const char * const names[] = {
+ [ND_CMD_ARS_CAP] = "ars_cap",
+ [ND_CMD_ARS_START] = "ars_start",
+ [ND_CMD_ARS_QUERY] = "ars_query",
+ };
+
+ if (cmd < ARRAY_SIZE(names) && names[cmd])
+ return names[cmd];
+ return "unknown";
+}
+
+static inline const char *nd_dimm_cmd_name(unsigned cmd)
+{
+ static const char * const names[] = {
+ [ND_CMD_SMART] = "smart",
+ [ND_CMD_SMART_THRESHOLD] = "smart_thresh",
+ [ND_CMD_DIMM_FLAGS] = "flags",
+ [ND_CMD_GET_CONFIG_SIZE] = "get_size",
+ [ND_CMD_GET_CONFIG_DATA] = "get_data",
+ [ND_CMD_SET_CONFIG_DATA] = "set_data",
+ [ND_CMD_VENDOR_EFFECT_LOG_SIZE] = "effect_size",
+ [ND_CMD_VENDOR_EFFECT_LOG] = "effect_log",
+ [ND_CMD_VENDOR] = "vendor",
+ };
+
+ if (cmd < ARRAY_SIZE(names) && names[cmd])
+ return names[cmd];
+ return "unknown";
+}
+
+#define ND_IOCTL 'N'
+
+#define ND_IOCTL_SMART _IOWR(ND_IOCTL, ND_CMD_SMART,\
+ struct nd_cmd_smart)
+
+#define ND_IOCTL_SMART_THRESHOLD _IOWR(ND_IOCTL, ND_CMD_SMART_THRESHOLD,\
+ struct nd_cmd_smart_threshold)
+
+#define ND_IOCTL_DIMM_FLAGS _IOWR(ND_IOCTL, ND_CMD_DIMM_FLAGS,\
+ struct nd_cmd_dimm_flags)
+
+#define ND_IOCTL_GET_CONFIG_SIZE _IOWR(ND_IOCTL, ND_CMD_GET_CONFIG_SIZE,\
+ struct nd_cmd_get_config_size)
+
+#define ND_IOCTL_GET_CONFIG_DATA _IOWR(ND_IOCTL, ND_CMD_GET_CONFIG_DATA,\
+ struct nd_cmd_get_config_data_hdr)
+
+#define ND_IOCTL_SET_CONFIG_DATA _IOWR(ND_IOCTL, ND_CMD_SET_CONFIG_DATA,\
+ struct nd_cmd_set_config_hdr)
+
+#define ND_IOCTL_VENDOR _IOWR(ND_IOCTL, ND_CMD_VENDOR,\
+ struct nd_cmd_vendor_hdr)
+
+#define ND_IOCTL_ARS_CAP _IOWR(ND_IOCTL, ND_CMD_ARS_CAP,\
+ struct nd_cmd_ars_cap)
+
+#define ND_IOCTL_ARS_START _IOWR(ND_IOCTL, ND_CMD_ARS_START,\
+ struct nd_cmd_ars_start)
+
+#define ND_IOCTL_ARS_QUERY _IOWR(ND_IOCTL, ND_CMD_ARS_QUERY,\
+ struct nd_cmd_ars_query)
+
+#endif /* __NDCTL_H__ */
* Implement the device-model infrastructure for loading modules and
attaching drivers to nd devices. This is a simple association of a
nd-device-type number with a driver that has a bitmask of supported
device types. To facilitate userspace bind/unbind operations 'modalias'
and 'devtype', that also appear in the uevent, are added as generic
sysfs attributes for all nd devices. The reason for the device-type
number is to support sub-types within a given parent devtype, be it a
vendor-specific sub-type or otherwise.
* The first consumer of this infrastructure is the driver
for dimm devices. It simply uses control messages to retrieve and
store the configuration-data image (label set) from each dimm.
Note: nd_device_register() arranges for asynchronous registration of
nd bus devices by default.
Cc: Greg KH <[email protected]>
Cc: Neil Brown <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/Makefile | 1
drivers/block/nd/acpi.c | 13 ++-
drivers/block/nd/bus.c | 168 +++++++++++++++++++++++++++++++++++++++++
drivers/block/nd/core.c | 43 ++++++++++
drivers/block/nd/dimm.c | 93 +++++++++++++++++++++++
drivers/block/nd/dimm_devs.c | 136 ++++++++++++++++++++++++++++++++-
drivers/block/nd/libnd.h | 2
drivers/block/nd/nd-private.h | 8 +-
drivers/block/nd/nd.h | 34 ++++++++
include/linux/nd.h | 39 ++++++++++
include/uapi/linux/ndctl.h | 6 +
11 files changed, 528 insertions(+), 15 deletions(-)
create mode 100644 drivers/block/nd/dimm.c
create mode 100644 drivers/block/nd/nd.h
create mode 100644 include/linux/nd.h
diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
index 35e4c1a7a8ff..842ba13253fd 100644
--- a/drivers/block/nd/Makefile
+++ b/drivers/block/nd/Makefile
@@ -21,3 +21,4 @@ nd_acpi-y := acpi.o
libnd-y := core.o
libnd-y += bus.o
libnd-y += dimm_devs.o
+libnd-y += dimm.o
diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
index c46e166695f7..bb0c2c764e78 100644
--- a/drivers/block/nd/acpi.c
+++ b/drivers/block/nd/acpi.c
@@ -22,6 +22,10 @@ static bool warn_checksum;
module_param(warn_checksum, bool, S_IRUGO|S_IWUSR);
MODULE_PARM_DESC(warn_checksum, "Turn checksum errors into warnings");
+static bool force_enable_dimms;
+module_param(force_enable_dimms, bool, S_IRUGO|S_IWUSR);
+MODULE_PARM_DESC(force_enable_dimms, "Ignore _STA (ACPI DIMM device) status");
+
enum {
NFIT_ACPI_NOTIFY_TABLE = 0x80,
};
@@ -627,6 +631,7 @@ static struct attribute_group nd_acpi_dimm_attribute_group = {
static const struct attribute_group *nd_acpi_dimm_attribute_groups[] = {
&nd_dimm_attribute_group,
+ &nd_device_attribute_group,
&nd_acpi_dimm_attribute_group,
NULL,
};
@@ -663,7 +668,7 @@ static int nd_acpi_add_dimm(struct acpi_nfit_desc *acpi_desc,
if (!adev_dimm) {
dev_err(dev, "no ACPI.NFIT device with _ADR %#x, disabling...\n",
nfit_handle);
- return -ENODEV;
+ return force_enable_dimms ? 0 : -ENODEV;
}
status = acpi_evaluate_integer(adev_dimm->handle, "_STA", NULL, &sta);
@@ -684,12 +689,13 @@ static int nd_acpi_add_dimm(struct acpi_nfit_desc *acpi_desc,
if (acpi_check_dsm(adev_dimm->handle, uuid, 1, 1ULL << i))
set_bit(i, &nfit_mem->dsm_mask);
- return rc;
+ return force_enable_dimms ? 0 : rc;
}
static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
{
struct nfit_mem *nfit_mem;
+ int dimm_count = 0;
list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) {
struct nd_dimm *nd_dimm;
@@ -723,9 +729,10 @@ static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
return -ENOMEM;
nfit_mem->nd_dimm = nd_dimm;
+ dimm_count++;
}
- return 0;
+ return nd_bus_validate_dimm_count(acpi_desc->nd_bus, dimm_count);
}
static void nd_acpi_init_dsms(struct acpi_nfit_desc *acpi_desc)
diff --git a/drivers/block/nd/bus.c b/drivers/block/nd/bus.c
index a271e01af4a9..7bd79f30e5e7 100644
--- a/drivers/block/nd/bus.c
+++ b/drivers/block/nd/bus.c
@@ -16,19 +16,183 @@
#include <linux/fcntl.h>
#include <linux/async.h>
#include <linux/ndctl.h>
+#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/fs.h>
#include <linux/io.h>
#include <linux/mm.h>
+#include <linux/nd.h>
#include "nd-private.h"
+#include "nd.h"
int nd_dimm_major;
static int nd_bus_major;
static struct class *nd_class;
-struct bus_type nd_bus_type = {
+static int to_nd_device_type(struct device *dev)
+{
+ if (is_nd_dimm(dev))
+ return ND_DEVICE_DIMM;
+
+ return 0;
+}
+
+static int nd_bus_uevent(struct device *dev, struct kobj_uevent_env *env)
+{
+ return add_uevent_var(env, "MODALIAS=" ND_DEVICE_MODALIAS_FMT,
+ to_nd_device_type(dev));
+}
+
+static int nd_bus_match(struct device *dev, struct device_driver *drv)
+{
+ struct nd_device_driver *nd_drv = to_nd_device_driver(drv);
+
+ return test_bit(to_nd_device_type(dev), &nd_drv->type);
+}
+
+static int nd_bus_probe(struct device *dev)
+{
+ struct nd_device_driver *nd_drv = to_nd_device_driver(dev->driver);
+ struct nd_bus *nd_bus = walk_to_nd_bus(dev);
+ int rc;
+
+ rc = nd_drv->probe(dev);
+ dev_dbg(&nd_bus->dev, "%s.probe(%s) = %d\n", dev->driver->name,
+ dev_name(dev), rc);
+ return rc;
+}
+
+static int nd_bus_remove(struct device *dev)
+{
+ struct nd_device_driver *nd_drv = to_nd_device_driver(dev->driver);
+ struct nd_bus *nd_bus = walk_to_nd_bus(dev);
+ int rc;
+
+ rc = nd_drv->remove(dev);
+ dev_dbg(&nd_bus->dev, "%s.remove(%s) = %d\n", dev->driver->name,
+ dev_name(dev), rc);
+ return rc;
+}
+
+static struct bus_type nd_bus_type = {
.name = "nd",
+ .uevent = nd_bus_uevent,
+ .match = nd_bus_match,
+ .probe = nd_bus_probe,
+ .remove = nd_bus_remove,
+};
+
+static ASYNC_DOMAIN_EXCLUSIVE(nd_async_domain);
+
+void nd_synchronize(void)
+{
+ async_synchronize_full_domain(&nd_async_domain);
+}
+EXPORT_SYMBOL_GPL(nd_synchronize);
+
+static void nd_async_device_register(void *d, async_cookie_t cookie)
+{
+ struct device *dev = d;
+
+ if (device_add(dev) != 0) {
+ dev_err(dev, "%s: failed\n", __func__);
+ put_device(dev);
+ }
+ put_device(dev);
+}
+
+static void nd_async_device_unregister(void *d, async_cookie_t cookie)
+{
+ struct device *dev = d;
+
+ device_unregister(dev);
+ put_device(dev);
+}
+
+void nd_device_register(struct device *dev)
+{
+ dev->bus = &nd_bus_type;
+ device_initialize(dev);
+ get_device(dev);
+ async_schedule_domain(nd_async_device_register, dev,
+ &nd_async_domain);
+}
+EXPORT_SYMBOL(nd_device_register);
+
+void nd_device_unregister(struct device *dev, enum nd_async_mode mode)
+{
+ switch (mode) {
+ case ND_ASYNC:
+ get_device(dev);
+ async_schedule_domain(nd_async_device_unregister, dev,
+ &nd_async_domain);
+ break;
+ case ND_SYNC:
+ nd_synchronize();
+ device_unregister(dev);
+ break;
+ }
+}
+EXPORT_SYMBOL(nd_device_unregister);
+
+/**
+ * __nd_driver_register() - register a region or a namespace driver
+ * @nd_drv: driver to register
+ * @owner: automatically set by nd_driver_register() macro
+ * @mod_name: automatically set by nd_driver_register() macro
+ */
+int __nd_driver_register(struct nd_device_driver *nd_drv, struct module *owner,
+ const char *mod_name)
+{
+ struct device_driver *drv = &nd_drv->drv;
+
+ if (!nd_drv->type) {
+ pr_debug("driver type bitmask not set (%pf)\n",
+ __builtin_return_address(0));
+ return -EINVAL;
+ }
+
+ if (!nd_drv->probe || !nd_drv->remove) {
+ pr_debug("->probe() and ->remove() must be specified\n");
+ return -EINVAL;
+ }
+
+ drv->bus = &nd_bus_type;
+ drv->owner = owner;
+ drv->mod_name = mod_name;
+
+ return driver_register(drv);
+}
+EXPORT_SYMBOL(__nd_driver_register);
+
+static ssize_t modalias_show(struct device *dev, struct device_attribute *attr,
+ char *buf)
+{
+ return sprintf(buf, ND_DEVICE_MODALIAS_FMT "\n",
+ to_nd_device_type(dev));
+}
+static DEVICE_ATTR_RO(modalias);
+
+static ssize_t devtype_show(struct device *dev, struct device_attribute *attr,
+ char *buf)
+{
+ return sprintf(buf, "%s\n", dev->type->name);
+}
+DEVICE_ATTR_RO(devtype);
+
+static struct attribute *nd_device_attributes[] = {
+ &dev_attr_modalias.attr,
+ &dev_attr_devtype.attr,
+ NULL,
+};
+
+/**
+ * nd_device_attribute_group - generic attributes for all devices on an nd bus
+ */
+struct attribute_group nd_device_attribute_group = {
+ .attrs = nd_device_attributes,
};
+EXPORT_SYMBOL_GPL(nd_device_attribute_group);
int nd_bus_create_ndctl(struct nd_bus *nd_bus)
{
@@ -402,7 +566,7 @@ int __init nd_bus_init(void)
return rc;
}
-void __exit nd_bus_exit(void)
+void nd_bus_exit(void)
{
class_destroy(nd_class);
unregister_chrdev(nd_bus_major, "ndctl");
diff --git a/drivers/block/nd/core.c b/drivers/block/nd/core.c
index b14bf47dc292..d8d1c9cb3f16 100644
--- a/drivers/block/nd/core.c
+++ b/drivers/block/nd/core.c
@@ -18,6 +18,7 @@
#include <linux/slab.h>
#include "nd-private.h"
#include "libnd.h"
+#include "nd.h"
LIST_HEAD(nd_bus_list);
DEFINE_MUTEX(nd_bus_list_mutex);
@@ -96,8 +97,33 @@ static ssize_t provider_show(struct device *dev,
}
static DEVICE_ATTR_RO(provider);
+static int flush_namespaces(struct device *dev, void *data)
+{
+ device_lock(dev);
+ device_unlock(dev);
+ return 0;
+}
+
+static int flush_regions_dimms(struct device *dev, void *data)
+{
+ device_lock(dev);
+ device_unlock(dev);
+ device_for_each_child(dev, NULL, flush_namespaces);
+ return 0;
+}
+
+static ssize_t wait_probe_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ nd_synchronize();
+ device_for_each_child(dev, NULL, flush_regions_dimms);
+ return sprintf(buf, "1\n");
+}
+static DEVICE_ATTR_RO(wait_probe);
+
static struct attribute *nd_bus_attributes[] = {
&dev_attr_commands.attr,
+ &dev_attr_wait_probe.attr,
&dev_attr_provider.attr,
NULL,
};
@@ -158,7 +184,7 @@ static int child_unregister(struct device *dev, void *data)
if (dev->class)
/* pass */;
else
- device_unregister(dev);
+ nd_device_unregister(dev, ND_SYNC);
return 0;
}
@@ -171,6 +197,7 @@ void nd_bus_unregister(struct nd_bus *nd_bus)
list_del_init(&nd_bus->list);
mutex_unlock(&nd_bus_list_mutex);
+ nd_synchronize();
device_for_each_child(&nd_bus->dev, NULL, child_unregister);
nd_bus_destroy_ndctl(nd_bus);
@@ -180,12 +207,24 @@ EXPORT_SYMBOL_GPL(nd_bus_unregister);
static __init int libnd_init(void)
{
- return nd_bus_init();
+ int rc;
+
+ rc = nd_bus_init();
+ if (rc)
+ return rc;
+ rc = nd_dimm_init();
+ if (rc)
+ goto err_dimm;
+ return 0;
+ err_dimm:
+ nd_bus_exit();
+ return rc;
}
static __exit void libnd_exit(void)
{
WARN_ON(!list_empty(&nd_bus_list));
+ nd_dimm_exit();
nd_bus_exit();
}
diff --git a/drivers/block/nd/dimm.c b/drivers/block/nd/dimm.c
new file mode 100644
index 000000000000..a4c8e3ffe97c
--- /dev/null
+++ b/drivers/block/nd/dimm.c
@@ -0,0 +1,93 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#include <linux/vmalloc.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/sizes.h>
+#include <linux/ndctl.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+#include <linux/nd.h>
+#include "nd.h"
+
+static void free_data(struct nd_dimm_drvdata *ndd)
+{
+ if (!ndd)
+ return;
+
+ if (ndd->data && is_vmalloc_addr(ndd->data))
+ vfree(ndd->data);
+ else
+ kfree(ndd->data);
+ kfree(ndd);
+}
+
+static int nd_dimm_probe(struct device *dev)
+{
+ struct nd_dimm_drvdata *ndd;
+ int rc;
+
+ ndd = kzalloc(sizeof(*ndd), GFP_KERNEL);
+ if (!ndd)
+ return -ENOMEM;
+
+ dev_set_drvdata(dev, ndd);
+ ndd->dev = dev;
+
+ rc = nd_dimm_init_nsarea(ndd);
+ if (rc)
+ goto err;
+
+ rc = nd_dimm_init_config_data(ndd);
+ if (rc)
+ goto err;
+
+ dev_dbg(dev, "config data size: %d\n", ndd->nsarea.config_size);
+
+ return 0;
+
+ err:
+ free_data(ndd);
+ return rc;
+
+}
+
+static int nd_dimm_remove(struct device *dev)
+{
+ struct nd_dimm_drvdata *ndd = dev_get_drvdata(dev);
+
+ free_data(ndd);
+
+ return 0;
+}
+
+static struct nd_device_driver nd_dimm_driver = {
+ .probe = nd_dimm_probe,
+ .remove = nd_dimm_remove,
+ .drv = {
+ .name = "nd_dimm",
+ },
+ .type = ND_DRIVER_DIMM,
+};
+
+int __init nd_dimm_init(void)
+{
+ return nd_driver_register(&nd_dimm_driver);
+}
+
+void __exit nd_dimm_exit(void)
+{
+ driver_unregister(&nd_dimm_driver.drv);
+}
+
+MODULE_ALIAS_ND_DEVICE(ND_DEVICE_DIMM);
diff --git a/drivers/block/nd/dimm_devs.c b/drivers/block/nd/dimm_devs.c
index 3fa26f61c3db..33b6d5336096 100644
--- a/drivers/block/nd/dimm_devs.c
+++ b/drivers/block/nd/dimm_devs.c
@@ -11,6 +11,7 @@
* General Public License for more details.
*/
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+#include <linux/vmalloc.h>
#include <linux/device.h>
#include <linux/ndctl.h>
#include <linux/slab.h>
@@ -18,9 +19,115 @@
#include <linux/fs.h>
#include <linux/mm.h>
#include "nd-private.h"
+#include "nd.h"
static DEFINE_IDA(dimm_ida);
+/*
+ * Retrieve bus and dimm handle and return if this bus supports
+ * get_config_data commands
+ */
+static int __validate_dimm(struct nd_dimm_drvdata *ndd)
+{
+ struct nd_dimm *nd_dimm;
+
+ if (!ndd)
+ return -EINVAL;
+
+ nd_dimm = to_nd_dimm(ndd->dev);
+
+ if (!nd_dimm->dsm_mask)
+ return -ENXIO;
+ if (!test_bit(ND_CMD_GET_CONFIG_DATA, nd_dimm->dsm_mask))
+ return -ENXIO;
+
+ return 0;
+}
+
+static int validate_dimm(struct nd_dimm_drvdata *ndd)
+{
+ int rc = __validate_dimm(ndd);
+
+ if (rc && ndd)
+ dev_dbg(ndd->dev, "%pf: %s error: %d\n",
+ __builtin_return_address(0), __func__, rc);
+ return rc;
+}
+
+/**
+ * nd_dimm_init_nsarea - determine the geometry of a dimm's namespace area
+ * @nd_dimm: dimm to initialize
+ */
+int nd_dimm_init_nsarea(struct nd_dimm_drvdata *ndd)
+{
+ struct nd_cmd_get_config_size *cmd = &ndd->nsarea;
+ struct nd_bus *nd_bus = walk_to_nd_bus(ndd->dev);
+ struct nd_bus_descriptor *nd_desc;
+ int rc = validate_dimm(ndd);
+
+ if (rc)
+ return rc;
+
+ if (cmd->config_size)
+ return 0; /* already valid */
+
+ memset(cmd, 0, sizeof(*cmd));
+ nd_desc = nd_bus->nd_desc;
+ return nd_desc->ndctl(nd_desc, to_nd_dimm(ndd->dev),
+ ND_CMD_GET_CONFIG_SIZE, cmd, sizeof(*cmd));
+}
+
+int nd_dimm_init_config_data(struct nd_dimm_drvdata *ndd)
+{
+ struct nd_bus *nd_bus = walk_to_nd_bus(ndd->dev);
+ struct nd_cmd_get_config_data_hdr *cmd;
+ struct nd_bus_descriptor *nd_desc;
+ int rc = validate_dimm(ndd);
+ u32 max_cmd_size, config_size;
+ size_t offset;
+
+ if (rc)
+ return rc;
+
+ if (ndd->data)
+ return 0;
+
+ if (ndd->nsarea.status || ndd->nsarea.max_xfer == 0)
+ return -ENXIO;
+
+ ndd->data = kmalloc(ndd->nsarea.config_size, GFP_KERNEL);
+ if (!ndd->data)
+ ndd->data = vmalloc(ndd->nsarea.config_size);
+
+ if (!ndd->data)
+ return -ENOMEM;
+
+ max_cmd_size = min_t(u32, PAGE_SIZE, ndd->nsarea.max_xfer);
+ cmd = kzalloc(max_cmd_size + sizeof(*cmd), GFP_KERNEL);
+ if (!cmd)
+ return -ENOMEM;
+
+ nd_desc = nd_bus->nd_desc;
+ for (config_size = ndd->nsarea.config_size, offset = 0;
+ config_size; config_size -= cmd->in_length,
+ offset += cmd->in_length) {
+ cmd->in_length = min(config_size, max_cmd_size);
+ cmd->in_offset = offset;
+ rc = nd_desc->ndctl(nd_desc, to_nd_dimm(ndd->dev),
+ ND_CMD_GET_CONFIG_DATA, cmd,
+ cmd->in_length + sizeof(*cmd));
+ if (rc || cmd->status) {
+ rc = -ENXIO;
+ break;
+ }
+ memcpy(ndd->data + offset, cmd->out_buf, cmd->in_length);
+ }
+ dev_dbg(ndd->dev, "%s: len: %zd rc: %d\n", __func__, offset, rc);
+ kfree(cmd);
+
+ return rc;
+}
+
static void nd_dimm_release(struct device *dev)
{
struct nd_dimm *nd_dimm = to_nd_dimm(dev);
@@ -111,14 +218,33 @@ struct nd_dimm *nd_dimm_create(struct nd_bus *nd_bus, void *provider_data,
dev_set_name(dev, "nmem%d", nd_dimm->id);
dev->parent = &nd_bus->dev;
dev->type = &nd_dimm_device_type;
- dev->bus = &nd_bus_type;
dev->devt = MKDEV(nd_dimm_major, nd_dimm->id);
dev->groups = groups;
- if (device_register(dev) != 0) {
- put_device(dev);
- return NULL;
- }
+ nd_device_register(dev);
return nd_dimm;
}
EXPORT_SYMBOL_GPL(nd_dimm_create);
+
+static int count_dimms(struct device *dev, void *c)
+{
+ int *count = c;
+
+ if (is_nd_dimm(dev))
+ (*count)++;
+ return 0;
+}
+
+int nd_bus_validate_dimm_count(struct nd_bus *nd_bus, int dimm_count)
+{
+ int count = 0;
+ /* Flush any possible dimm registration failures */
+ nd_synchronize();
+
+ device_for_each_child(&nd_bus->dev, &count, count_dimms);
+ dev_dbg(&nd_bus->dev, "%s: count: %d\n", __func__, count);
+ if (count != dimm_count)
+ return -ENXIO;
+ return 0;
+}
+EXPORT_SYMBOL_GPL(nd_bus_validate_dimm_count);
diff --git a/drivers/block/nd/libnd.h b/drivers/block/nd/libnd.h
index 23dae8c1e65b..3b91df914b76 100644
--- a/drivers/block/nd/libnd.h
+++ b/drivers/block/nd/libnd.h
@@ -29,6 +29,7 @@ enum {
extern struct attribute_group nd_bus_attribute_group;
extern struct attribute_group nd_dimm_attribute_group;
+extern struct attribute_group nd_device_attribute_group;
struct nd_dimm;
struct nd_bus_descriptor;
@@ -69,4 +70,5 @@ u32 nd_cmd_in_size(struct nd_dimm *nd_dimm, int cmd,
u32 nd_cmd_out_size(struct nd_dimm *nd_dimm, int cmd,
const struct nd_cmd_desc *desc, int idx, const u32 *in_field,
const u32 *out_field);
+int nd_bus_validate_dimm_count(struct nd_bus *nd_bus, int dimm_count);
#endif /* __LIBND_H__ */
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
index 968e5089f72c..35ab0d476d15 100644
--- a/drivers/block/nd/nd-private.h
+++ b/drivers/block/nd/nd-private.h
@@ -17,7 +17,6 @@
extern struct list_head nd_bus_list;
extern struct mutex nd_bus_list_mutex;
-extern struct bus_type nd_bus_type;
extern int nd_dimm_major;
struct nd_bus {
@@ -35,10 +34,13 @@ struct nd_dimm {
int id;
};
-bool is_nd_dimm(struct device *dev);
struct nd_bus *walk_to_nd_bus(struct device *nd_dev);
int __init nd_bus_init(void);
-void __exit nd_bus_exit(void);
+void nd_bus_exit(void);
+int __init nd_dimm_init(void);
+void __exit nd_dimm_exit(void);
int nd_bus_create_ndctl(struct nd_bus *nd_bus);
void nd_bus_destroy_ndctl(struct nd_bus *nd_bus);
+void nd_synchronize(void);
+bool is_nd_dimm(struct device *dev);
#endif /* __ND_PRIVATE_H__ */
diff --git a/drivers/block/nd/nd.h b/drivers/block/nd/nd.h
new file mode 100644
index 000000000000..1a5a081ce640
--- /dev/null
+++ b/drivers/block/nd/nd.h
@@ -0,0 +1,34 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#ifndef __ND_H__
+#define __ND_H__
+#include <linux/device.h>
+#include <linux/mutex.h>
+#include <linux/ndctl.h>
+
+struct nd_dimm_drvdata {
+ struct device *dev;
+ struct nd_cmd_get_config_size nsarea;
+ void *data;
+};
+
+enum nd_async_mode {
+ ND_SYNC,
+ ND_ASYNC,
+};
+
+void nd_device_register(struct device *dev);
+void nd_device_unregister(struct device *dev, enum nd_async_mode mode);
+int nd_dimm_init_nsarea(struct nd_dimm_drvdata *ndd);
+int nd_dimm_init_config_data(struct nd_dimm_drvdata *ndd);
+#endif /* __ND_H__ */
diff --git a/include/linux/nd.h b/include/linux/nd.h
new file mode 100644
index 000000000000..e074f67e53a3
--- /dev/null
+++ b/include/linux/nd.h
@@ -0,0 +1,39 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#ifndef __LINUX_ND_H__
+#define __LINUX_ND_H__
+#include <linux/ndctl.h>
+#include <linux/device.h>
+
+struct nd_device_driver {
+ struct device_driver drv;
+ unsigned long type;
+ int (*probe)(struct device *dev);
+ int (*remove)(struct device *dev);
+};
+
+static inline struct nd_device_driver *to_nd_device_driver(
+ struct device_driver *drv)
+{
+ return container_of(drv, struct nd_device_driver, drv);
+}
+
+#define MODULE_ALIAS_ND_DEVICE(type) \
+ MODULE_ALIAS("nd:t" __stringify(type) "*")
+#define ND_DEVICE_MODALIAS_FMT "nd:t%d"
+
+int __must_check __nd_driver_register(struct nd_device_driver *nd_drv,
+ struct module *module, const char *mod_name);
+#define nd_driver_register(driver) \
+ __nd_driver_register(driver, THIS_MODULE, KBUILD_MODNAME)
+#endif /* __LINUX_ND_H__ */
diff --git a/include/uapi/linux/ndctl.h b/include/uapi/linux/ndctl.h
index 62c01bf76198..1ccd2c633193 100644
--- a/include/uapi/linux/ndctl.h
+++ b/include/uapi/linux/ndctl.h
@@ -175,4 +175,10 @@ static inline const char *nd_dimm_cmd_name(unsigned cmd)
#define ND_IOCTL_ARS_QUERY _IOWR(ND_IOCTL, ND_CMD_ARS_QUERY,\
struct nd_cmd_ars_query)
+
+#define ND_DEVICE_DIMM 1 /* nd_dimm: container for "config data" */
+
+enum nd_driver_flags {
+ ND_DRIVER_DIMM = 1 << ND_DEVICE_DIMM,
+};
#endif /* __NDCTL_H__ */
A "region" device represents the maximum capacity of a BLK range (mmio
block-data-window(s)), or a PMEM range (DAX-capable persistent memory or
volatile memory), without regard for aliasing. Aliasing, in the
dimm-local address space (DPA), is resolved by metadata on a dimm to
designate which exclusive interface will access the aliased DPA ranges.
Support for the per-dimm metadata/label arrvies is in a subsequent
patch.
The name format of "region" devices is "regionN" where, like dimms, N is
a global ida index assigned at discovery time. This id is not reliable
across reboots nor in the presence of hotplug. Look to attributes of
the region or static id-data of the sub-namespace to generate a
persistent name.
"region"s have 2 generic attributes "size", and "mapping"s where:
- size: the block-data-window accessible capacity or the span of the
spa-range in the case of pm.
- mappingN: a tuple describing a dimm's contribution to the region's
capacity in the format (<nmemX>,<dpa>,<size>). For a
PMEM-region there will be at least one mapping per dimm in the interleave
set. For a BLK-region there is only "mapping0" listing the starting dimm
offset of the block-data-window and the available capacity of that
window (matches "size" above).
The max number of mappings per "region" is hard coded per the constraints of
sysfs attribute groups. That said the number of mappings per region should
never exceed the maximum number of possible dimms in the system. If the
current number turns out to not be enough then the "mappings" attribute
clarifies how many there are supposed to be. "32 should be enough for
anybody...".
Cc: Neil Brown <[email protected]>
Cc: <[email protected]>
Cc: Greg KH <[email protected]>
Cc: Robert Moore <[email protected]>
Cc: Rafael J. Wysocki <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/Makefile | 1
drivers/block/nd/acpi.c | 130 ++++++++++++++++++
drivers/block/nd/libnd.h | 25 +++
drivers/block/nd/nd-private.h | 3
drivers/block/nd/nd.h | 11 +
drivers/block/nd/region_devs.c | 294 ++++++++++++++++++++++++++++++++++++++++
6 files changed, 463 insertions(+), 1 deletion(-)
create mode 100644 drivers/block/nd/region_devs.c
diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
index 842ba13253fd..6010469c4d4c 100644
--- a/drivers/block/nd/Makefile
+++ b/drivers/block/nd/Makefile
@@ -22,3 +22,4 @@ libnd-y := core.o
libnd-y += bus.o
libnd-y += dimm_devs.o
libnd-y += dimm.o
+libnd-y += region_devs.o
diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
index bb0c2c764e78..41d0bb732b3e 100644
--- a/drivers/block/nd/acpi.c
+++ b/drivers/block/nd/acpi.c
@@ -751,12 +751,136 @@ static void nd_acpi_init_dsms(struct acpi_nfit_desc *acpi_desc)
set_bit(i, &nd_desc->dsm_mask);
}
+static ssize_t spa_index_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_region *nd_region = to_nd_region(dev);
+ struct nfit_spa *nfit_spa = nd_region_provider_data(nd_region);
+
+ return sprintf(buf, "%d\n", nfit_spa->spa->spa_index);
+}
+static DEVICE_ATTR_RO(spa_index);
+
+static struct attribute *nd_acpi_region_attributes[] = {
+ &dev_attr_spa_index.attr,
+ NULL,
+};
+
+static struct attribute_group nd_acpi_region_attribute_group = {
+ .name = "nfit",
+ .attrs = nd_acpi_region_attributes,
+};
+
+static const struct attribute_group *nd_acpi_region_attribute_groups[] = {
+ &nd_region_attribute_group,
+ &nd_mapping_attribute_group,
+ &nd_acpi_region_attribute_group,
+ NULL,
+};
+
+static int nd_acpi_register_region(struct acpi_nfit_desc *acpi_desc,
+ struct nfit_spa *nfit_spa)
+{
+ static struct nd_mapping nd_mappings[ND_MAX_MAPPINGS];
+ struct acpi_nfit_spa *spa = nfit_spa->spa;
+ struct nfit_memdev *nfit_memdev;
+ struct nd_region_desc ndr_desc;
+ int spa_type, count = 0;
+ struct resource res;
+ u16 spa_index;
+
+ spa_type = nfit_spa_type(spa);
+ spa_index = spa->spa_index;
+ if (spa_index == 0) {
+ dev_dbg(acpi_desc->dev, "%s: detected invalid spa index\n",
+ __func__);
+ return 0;
+ }
+
+ memset(&res, 0, sizeof(res));
+ memset(&nd_mappings, 0, sizeof(nd_mappings));
+ memset(&ndr_desc, 0, sizeof(ndr_desc));
+ res.start = spa->spa_base;
+ res.end = res.start + spa->spa_length - 1;
+ ndr_desc.res = &res;
+ ndr_desc.provider_data = nfit_spa;
+ ndr_desc.attr_groups = nd_acpi_region_attribute_groups;
+ list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) {
+ struct acpi_nfit_memdev *memdev = nfit_memdev->memdev;
+ struct nd_mapping *nd_mapping;
+ struct nd_dimm *nd_dimm;
+
+ if (memdev->spa_index != spa_index)
+ continue;
+ if (count >= ND_MAX_MAPPINGS) {
+ dev_err(acpi_desc->dev, "spa%d exceeds max mappings %d\n",
+ spa_index, ND_MAX_MAPPINGS);
+ return -ENXIO;
+ }
+ nd_dimm = nd_acpi_dimm_by_handle(acpi_desc, memdev->nfit_handle);
+ if (!nd_dimm) {
+ dev_err(acpi_desc->dev, "spa%d dimm: %#x not found\n",
+ spa_index, memdev->nfit_handle);
+ return -ENODEV;
+ }
+ nd_mapping = &nd_mappings[count++];
+ nd_mapping->nd_dimm = nd_dimm;
+ if (spa_type == NFIT_SPA_PM || spa_type == NFIT_SPA_VOLATILE) {
+ nd_mapping->start = memdev->region_dpa;
+ nd_mapping->size = memdev->region_len;
+ } else if (spa_type == NFIT_SPA_DCR) {
+ struct nfit_mem *nfit_mem;
+ int blk_valid = 1;
+
+ nfit_mem = nd_dimm_provider_data(nd_dimm);
+ if (!nfit_mem || !nfit_mem->bdw) {
+ dev_dbg(acpi_desc->dev, "%s: spa%d missing bdw\n",
+ nd_dimm_name(nd_dimm), spa_index);
+ blk_valid = 0;
+ } else {
+ nd_mapping->size = nfit_mem->bdw->blk_capacity;
+ nd_mapping->start = nfit_mem->bdw->blk_offset;
+ }
+
+ ndr_desc.nd_mapping = nd_mapping;
+ ndr_desc.num_mappings = blk_valid;
+ if (!nd_blk_region_create(acpi_desc->nd_bus, &ndr_desc))
+ return -ENOMEM;
+ }
+ }
+
+ ndr_desc.nd_mapping = nd_mappings;
+ ndr_desc.num_mappings = count;
+ if (spa_type == NFIT_SPA_PM) {
+ if (!nd_pmem_region_create(acpi_desc->nd_bus, &ndr_desc))
+ return -ENOMEM;
+ } else if (spa_type == NFIT_SPA_VOLATILE) {
+ if (!nd_volatile_region_create(acpi_desc->nd_bus, &ndr_desc))
+ return -ENOMEM;
+ }
+ return 0;
+}
+
+static int nd_acpi_register_regions(struct acpi_nfit_desc *acpi_desc)
+{
+ struct nfit_spa *nfit_spa;
+
+ list_for_each_entry(nfit_spa, &acpi_desc->spas, list) {
+ int rc = nd_acpi_register_region(acpi_desc, nfit_spa);
+
+ if (rc)
+ return rc;
+ }
+ return 0;
+}
+
int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
{
struct device *dev = acpi_desc->dev;
const void *end;
u8 *data, sum;
acpi_size i;
+ int rc;
INIT_LIST_HEAD(&acpi_desc->spas);
INIT_LIST_HEAD(&acpi_desc->dcrs);
@@ -790,7 +914,11 @@ int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
nd_acpi_init_dsms(acpi_desc);
- return nd_acpi_register_dimms(acpi_desc);
+ rc = nd_acpi_register_dimms(acpi_desc);
+ if (rc)
+ return rc;
+
+ return nd_acpi_register_regions(acpi_desc);
}
EXPORT_SYMBOL_GPL(nd_acpi_nfit_init);
diff --git a/drivers/block/nd/libnd.h b/drivers/block/nd/libnd.h
index 3b91df914b76..630a703d1316 100644
--- a/drivers/block/nd/libnd.h
+++ b/drivers/block/nd/libnd.h
@@ -25,11 +25,14 @@ enum {
ND_CMD_MAX_ELEM = 4,
ND_CMD_MAX_ENVELOPE = 16,
ND_CMD_ARS_QUERY_MAX = SZ_4K,
+ ND_MAX_MAPPINGS = 32,
};
extern struct attribute_group nd_bus_attribute_group;
extern struct attribute_group nd_dimm_attribute_group;
extern struct attribute_group nd_device_attribute_group;
+extern struct attribute_group nd_region_attribute_group;
+extern struct attribute_group nd_mapping_attribute_group;
struct nd_dimm;
struct nd_bus_descriptor;
@@ -37,6 +40,12 @@ typedef int (*ndctl_fn)(struct nd_bus_descriptor *nd_desc,
struct nd_dimm *nd_dimm, unsigned int cmd, void *buf,
unsigned int buf_len);
+struct nd_mapping {
+ struct nd_dimm *nd_dimm;
+ u64 start;
+ u64 size;
+};
+
struct nd_bus_descriptor {
const struct attribute_group **attr_groups;
unsigned long dsm_mask;
@@ -51,15 +60,25 @@ struct nd_cmd_desc {
int out_sizes[ND_CMD_MAX_ELEM];
};
+struct nd_region_desc {
+ struct resource *res;
+ struct nd_mapping *nd_mapping;
+ u16 num_mappings;
+ const struct attribute_group **attr_groups;
+ void *provider_data;
+};
+
struct nd_bus;
struct nd_bus *nd_bus_register(struct device *parent,
struct nd_bus_descriptor *nfit_desc);
void nd_bus_unregister(struct nd_bus *nd_bus);
struct nd_bus *to_nd_bus(struct device *dev);
struct nd_dimm *to_nd_dimm(struct device *dev);
+struct nd_region *to_nd_region(struct device *dev);
struct nd_bus_descriptor *to_nd_desc(struct nd_bus *nd_bus);
const char *nd_dimm_name(struct nd_dimm *nd_dimm);
void *nd_dimm_provider_data(struct nd_dimm *nd_dimm);
+void *nd_region_provider_data(struct nd_region *nd_region);
struct nd_dimm *nd_dimm_create(struct nd_bus *nd_bus, void *provider_data,
const struct attribute_group **groups, unsigned long flags,
unsigned long *dsm_mask);
@@ -71,4 +90,10 @@ u32 nd_cmd_out_size(struct nd_dimm *nd_dimm, int cmd,
const struct nd_cmd_desc *desc, int idx, const u32 *in_field,
const u32 *out_field);
int nd_bus_validate_dimm_count(struct nd_bus *nd_bus, int dimm_count);
+struct nd_region *nd_pmem_region_create(struct nd_bus *nd_bus,
+ struct nd_region_desc *ndr_desc);
+struct nd_region *nd_blk_region_create(struct nd_bus *nd_bus,
+ struct nd_region_desc *ndr_desc);
+struct nd_region *nd_volatile_region_create(struct nd_bus *nd_bus,
+ struct nd_region_desc *ndr_desc);
#endif /* __LIBND_H__ */
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
index 35ab0d476d15..838b6f958c00 100644
--- a/drivers/block/nd/nd-private.h
+++ b/drivers/block/nd/nd-private.h
@@ -42,5 +42,8 @@ void __exit nd_dimm_exit(void);
int nd_bus_create_ndctl(struct nd_bus *nd_bus);
void nd_bus_destroy_ndctl(struct nd_bus *nd_bus);
void nd_synchronize(void);
+int nd_bus_register_dimms(struct nd_bus *nd_bus);
+int nd_bus_register_regions(struct nd_bus *nd_bus);
+int nd_match_dimm(struct device *dev, void *data);
bool is_nd_dimm(struct device *dev);
#endif /* __ND_PRIVATE_H__ */
diff --git a/drivers/block/nd/nd.h b/drivers/block/nd/nd.h
index 1a5a081ce640..cae83de12c45 100644
--- a/drivers/block/nd/nd.h
+++ b/drivers/block/nd/nd.h
@@ -15,6 +15,7 @@
#include <linux/device.h>
#include <linux/mutex.h>
#include <linux/ndctl.h>
+#include "libnd.h"
struct nd_dimm_drvdata {
struct device *dev;
@@ -22,6 +23,16 @@ struct nd_dimm_drvdata {
void *data;
};
+struct nd_region {
+ struct device dev;
+ u16 ndr_mappings;
+ u64 ndr_size;
+ u64 ndr_start;
+ int id;
+ void *provider_data;
+ struct nd_mapping mapping[0];
+};
+
enum nd_async_mode {
ND_SYNC,
ND_ASYNC,
diff --git a/drivers/block/nd/region_devs.c b/drivers/block/nd/region_devs.c
new file mode 100644
index 000000000000..12a5415acfcc
--- /dev/null
+++ b/drivers/block/nd/region_devs.c
@@ -0,0 +1,294 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#include <linux/slab.h>
+#include <linux/io.h>
+#include "nd-private.h"
+#include "nd.h"
+
+static DEFINE_IDA(region_ida);
+
+static void nd_region_release(struct device *dev)
+{
+ struct nd_region *nd_region = to_nd_region(dev);
+ u16 i;
+
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ struct nd_dimm *nd_dimm = nd_mapping->nd_dimm;
+
+ put_device(&nd_dimm->dev);
+ }
+ ida_simple_remove(®ion_ida, nd_region->id);
+ kfree(nd_region);
+}
+
+static struct device_type nd_blk_device_type = {
+ .name = "nd_blk",
+ .release = nd_region_release,
+};
+
+static struct device_type nd_pmem_device_type = {
+ .name = "nd_pmem",
+ .release = nd_region_release,
+};
+
+static struct device_type nd_volatile_device_type = {
+ .name = "nd_volatile",
+ .release = nd_region_release,
+};
+
+static bool is_nd_pmem(struct device *dev)
+{
+ return dev ? dev->type == &nd_pmem_device_type : false;
+}
+
+struct nd_region *to_nd_region(struct device *dev)
+{
+ struct nd_region *nd_region = container_of(dev, struct nd_region, dev);
+
+ WARN_ON(dev->type->release != nd_region_release);
+ return nd_region;
+}
+EXPORT_SYMBOL_GPL(to_nd_region);
+
+static ssize_t size_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_region *nd_region = to_nd_region(dev);
+ unsigned long long size = 0;
+
+ if (is_nd_pmem(dev)) {
+ size = nd_region->ndr_size;
+ } else if (nd_region->ndr_mappings == 1) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[0];
+
+ size = nd_mapping->size;
+ }
+
+ return sprintf(buf, "%llu\n", size);
+}
+static DEVICE_ATTR_RO(size);
+
+static ssize_t mappings_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_region *nd_region = to_nd_region(dev);
+
+ return sprintf(buf, "%d\n", nd_region->ndr_mappings);
+}
+static DEVICE_ATTR_RO(mappings);
+
+static struct attribute *nd_region_attributes[] = {
+ &dev_attr_size.attr,
+ &dev_attr_mappings.attr,
+ NULL,
+};
+
+struct attribute_group nd_region_attribute_group = {
+ .attrs = nd_region_attributes,
+};
+EXPORT_SYMBOL_GPL(nd_region_attribute_group);
+
+static ssize_t mappingN(struct device *dev, char *buf, int n)
+{
+ struct nd_region *nd_region = to_nd_region(dev);
+ struct nd_mapping *nd_mapping;
+ struct nd_dimm *nd_dimm;
+
+ if (n >= nd_region->ndr_mappings)
+ return -ENXIO;
+ nd_mapping = &nd_region->mapping[n];
+ nd_dimm = nd_mapping->nd_dimm;
+
+ return sprintf(buf, "%s,%llu,%llu\n", dev_name(&nd_dimm->dev),
+ nd_mapping->start, nd_mapping->size);
+}
+
+#define REGION_MAPPING(idx) \
+static ssize_t mapping##idx##_show(struct device *dev, \
+ struct device_attribute *attr, char *buf) \
+{ \
+ return mappingN(dev, buf, idx); \
+} \
+static DEVICE_ATTR_RO(mapping##idx)
+
+/*
+ * 32 should be enough for a while, even in the presence of socket
+ * interleave a 32-way interleave set is a degenerate case.
+ */
+REGION_MAPPING(0);
+REGION_MAPPING(1);
+REGION_MAPPING(2);
+REGION_MAPPING(3);
+REGION_MAPPING(4);
+REGION_MAPPING(5);
+REGION_MAPPING(6);
+REGION_MAPPING(7);
+REGION_MAPPING(8);
+REGION_MAPPING(9);
+REGION_MAPPING(10);
+REGION_MAPPING(11);
+REGION_MAPPING(12);
+REGION_MAPPING(13);
+REGION_MAPPING(14);
+REGION_MAPPING(15);
+REGION_MAPPING(16);
+REGION_MAPPING(17);
+REGION_MAPPING(18);
+REGION_MAPPING(19);
+REGION_MAPPING(20);
+REGION_MAPPING(21);
+REGION_MAPPING(22);
+REGION_MAPPING(23);
+REGION_MAPPING(24);
+REGION_MAPPING(25);
+REGION_MAPPING(26);
+REGION_MAPPING(27);
+REGION_MAPPING(28);
+REGION_MAPPING(29);
+REGION_MAPPING(30);
+REGION_MAPPING(31);
+
+static umode_t nd_mapping_visible(struct kobject *kobj, struct attribute *a, int n)
+{
+ struct device *dev = container_of(kobj, struct device, kobj);
+ struct nd_region *nd_region = to_nd_region(dev);
+
+ if (n < nd_region->ndr_mappings)
+ return a->mode;
+ return 0;
+}
+
+static struct attribute *nd_mapping_attributes[] = {
+ &dev_attr_mapping0.attr,
+ &dev_attr_mapping1.attr,
+ &dev_attr_mapping2.attr,
+ &dev_attr_mapping3.attr,
+ &dev_attr_mapping4.attr,
+ &dev_attr_mapping5.attr,
+ &dev_attr_mapping6.attr,
+ &dev_attr_mapping7.attr,
+ &dev_attr_mapping8.attr,
+ &dev_attr_mapping9.attr,
+ &dev_attr_mapping10.attr,
+ &dev_attr_mapping11.attr,
+ &dev_attr_mapping12.attr,
+ &dev_attr_mapping13.attr,
+ &dev_attr_mapping14.attr,
+ &dev_attr_mapping15.attr,
+ &dev_attr_mapping16.attr,
+ &dev_attr_mapping17.attr,
+ &dev_attr_mapping18.attr,
+ &dev_attr_mapping19.attr,
+ &dev_attr_mapping20.attr,
+ &dev_attr_mapping21.attr,
+ &dev_attr_mapping22.attr,
+ &dev_attr_mapping23.attr,
+ &dev_attr_mapping24.attr,
+ &dev_attr_mapping25.attr,
+ &dev_attr_mapping26.attr,
+ &dev_attr_mapping27.attr,
+ &dev_attr_mapping28.attr,
+ &dev_attr_mapping29.attr,
+ &dev_attr_mapping30.attr,
+ &dev_attr_mapping31.attr,
+ NULL,
+};
+
+struct attribute_group nd_mapping_attribute_group = {
+ .is_visible = nd_mapping_visible,
+ .attrs = nd_mapping_attributes,
+};
+EXPORT_SYMBOL_GPL(nd_mapping_attribute_group);
+
+void *nd_region_provider_data(struct nd_region *nd_region)
+{
+ return nd_region->provider_data;
+}
+EXPORT_SYMBOL_GPL(nd_region_provider_data);
+
+static noinline struct nd_region *nd_region_create(struct nd_bus *nd_bus,
+ struct nd_region_desc *ndr_desc, struct device_type *dev_type)
+{
+ struct nd_region *nd_region;
+ struct device *dev;
+ u16 i;
+
+ for (i = 0; i < ndr_desc->num_mappings; i++) {
+ struct nd_mapping *nd_mapping = &ndr_desc->nd_mapping[i];
+ struct nd_dimm *nd_dimm = nd_mapping->nd_dimm;
+
+ if ((nd_mapping->start | nd_mapping->size) % SZ_4K) {
+ dev_err(&nd_bus->dev, "%pf: %s mapping%d is not 4K aligned\n",
+ __builtin_return_address(0),
+ dev_name(&nd_dimm->dev), i);
+
+ return NULL;
+ }
+ }
+
+ nd_region = kzalloc(sizeof(struct nd_region)
+ + sizeof(struct nd_mapping) * ndr_desc->num_mappings,
+ GFP_KERNEL);
+ if (!nd_region)
+ return NULL;
+ nd_region->id = ida_simple_get(®ion_ida, 0, 0, GFP_KERNEL);
+ if (nd_region->id < 0) {
+ kfree(nd_region);
+ return NULL;
+ }
+
+ memcpy(nd_region->mapping, ndr_desc->nd_mapping,
+ sizeof(struct nd_mapping) * ndr_desc->num_mappings);
+ for (i = 0; i < ndr_desc->num_mappings; i++) {
+ struct nd_mapping *nd_mapping = &ndr_desc->nd_mapping[i];
+ struct nd_dimm *nd_dimm = nd_mapping->nd_dimm;
+
+ get_device(&nd_dimm->dev);
+ }
+ nd_region->ndr_mappings = ndr_desc->num_mappings;
+ nd_region->provider_data = ndr_desc->provider_data;
+ dev = &nd_region->dev;
+ dev_set_name(dev, "region%d", nd_region->id);
+ dev->parent = &nd_bus->dev;
+ dev->type = dev_type;
+ dev->groups = ndr_desc->attr_groups;
+ nd_region->ndr_size = resource_size(ndr_desc->res);
+ nd_region->ndr_start = ndr_desc->res->start;
+ nd_device_register(dev);
+
+ return nd_region;
+}
+
+struct nd_region *nd_pmem_region_create(struct nd_bus *nd_bus,
+ struct nd_region_desc *ndr_desc)
+{
+ return nd_region_create(nd_bus, ndr_desc, &nd_pmem_device_type);
+}
+EXPORT_SYMBOL_GPL(nd_pmem_region_create);
+
+struct nd_region *nd_blk_region_create(struct nd_bus *nd_bus,
+ struct nd_region_desc *ndr_desc)
+{
+ if (ndr_desc->num_mappings > 1)
+ return NULL;
+ return nd_region_create(nd_bus, ndr_desc, &nd_blk_device_type);
+}
+EXPORT_SYMBOL_GPL(nd_blk_region_create);
+
+struct nd_region *nd_volatile_region_create(struct nd_bus *nd_bus,
+ struct nd_region_desc *ndr_desc)
+{
+ return nd_region_create(nd_bus, ndr_desc, &nd_volatile_device_type);
+}
+EXPORT_SYMBOL_GPL(nd_volatile_region_create);
The libnd region driver is an intermediary driver that translates
non-volatile "region"s into "namespace" sub-devices that are surfaced by
persistent memory block-device drivers (PMEM and BLK).
ACPI 6 introduces the concept that a given nvdimm may offer multiple
access modes to its media through either direct PMEM load/store access,
or windowed BLK mode. Existing nvdimms mostly implement a PMEM
interface, some offer a BLK-like mode, but never both. If an nvdimm is
single interfaced, then there is no need for dimm metadata labels. For
these devices we can take the region boundaries directly to create a
child namespace device (nd_namespace_io).
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/Makefile | 2 +
drivers/block/nd/acpi.c | 1
drivers/block/nd/bus.c | 26 +++++++++
drivers/block/nd/core.c | 13 +++-
drivers/block/nd/dimm.c | 2 -
drivers/block/nd/libnd.h | 6 +-
drivers/block/nd/namespace_devs.c | 111 +++++++++++++++++++++++++++++++++++++
drivers/block/nd/nd-private.h | 9 ++-
drivers/block/nd/nd.h | 8 +++
drivers/block/nd/region.c | 88 +++++++++++++++++++++++++++++
drivers/block/nd/region_devs.c | 61 ++++++++++++++++++++
include/linux/nd.h | 10 +++
include/uapi/linux/ndctl.h | 10 +++
13 files changed, 338 insertions(+), 9 deletions(-)
create mode 100644 drivers/block/nd/namespace_devs.c
create mode 100644 drivers/block/nd/region.c
diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
index 6010469c4d4c..0fb0891e1817 100644
--- a/drivers/block/nd/Makefile
+++ b/drivers/block/nd/Makefile
@@ -23,3 +23,5 @@ libnd-y += bus.o
libnd-y += dimm_devs.o
libnd-y += dimm.o
libnd-y += region_devs.o
+libnd-y += region.o
+libnd-y += namespace_devs.o
diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
index 41d0bb732b3e..c3dda74f73d7 100644
--- a/drivers/block/nd/acpi.c
+++ b/drivers/block/nd/acpi.c
@@ -774,6 +774,7 @@ static struct attribute_group nd_acpi_region_attribute_group = {
static const struct attribute_group *nd_acpi_region_attribute_groups[] = {
&nd_region_attribute_group,
&nd_mapping_attribute_group,
+ &nd_device_attribute_group,
&nd_acpi_region_attribute_group,
NULL,
};
diff --git a/drivers/block/nd/bus.c b/drivers/block/nd/bus.c
index 7bd79f30e5e7..46568d182559 100644
--- a/drivers/block/nd/bus.c
+++ b/drivers/block/nd/bus.c
@@ -13,6 +13,7 @@
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/vmalloc.h>
#include <linux/uaccess.h>
+#include <linux/module.h>
#include <linux/fcntl.h>
#include <linux/async.h>
#include <linux/ndctl.h>
@@ -33,6 +34,12 @@ static int to_nd_device_type(struct device *dev)
{
if (is_nd_dimm(dev))
return ND_DEVICE_DIMM;
+ else if (is_nd_pmem(dev))
+ return ND_DEVICE_REGION_PMEM;
+ else if (is_nd_blk(dev))
+ return ND_DEVICE_REGION_BLK;
+ else if (is_nd_pmem(dev->parent) || is_nd_blk(dev->parent))
+ return nd_region_to_namespace_type(to_nd_region(dev->parent));
return 0;
}
@@ -50,27 +57,46 @@ static int nd_bus_match(struct device *dev, struct device_driver *drv)
return test_bit(to_nd_device_type(dev), &nd_drv->type);
}
+static struct module *to_bus_provider(struct device *dev)
+{
+ /* pin bus providers while regions are enabled */
+ if (is_nd_pmem(dev) || is_nd_blk(dev)) {
+ struct nd_bus *nd_bus = walk_to_nd_bus(dev);
+
+ return nd_bus->module;
+ }
+ return NULL;
+}
+
static int nd_bus_probe(struct device *dev)
{
struct nd_device_driver *nd_drv = to_nd_device_driver(dev->driver);
+ struct module *provider = to_bus_provider(dev);
struct nd_bus *nd_bus = walk_to_nd_bus(dev);
int rc;
+ if (!try_module_get(provider))
+ return -ENXIO;
+
rc = nd_drv->probe(dev);
dev_dbg(&nd_bus->dev, "%s.probe(%s) = %d\n", dev->driver->name,
dev_name(dev), rc);
+ if (rc != 0)
+ module_put(provider);
return rc;
}
static int nd_bus_remove(struct device *dev)
{
struct nd_device_driver *nd_drv = to_nd_device_driver(dev->driver);
+ struct module *provider = to_bus_provider(dev);
struct nd_bus *nd_bus = walk_to_nd_bus(dev);
int rc;
rc = nd_drv->remove(dev);
dev_dbg(&nd_bus->dev, "%s.remove(%s) = %d\n", dev->driver->name,
dev_name(dev), rc);
+ module_put(provider);
return rc;
}
diff --git a/drivers/block/nd/core.c b/drivers/block/nd/core.c
index d8d1c9cb3f16..646e424ae36c 100644
--- a/drivers/block/nd/core.c
+++ b/drivers/block/nd/core.c
@@ -133,8 +133,8 @@ struct attribute_group nd_bus_attribute_group = {
};
EXPORT_SYMBOL_GPL(nd_bus_attribute_group);
-struct nd_bus *nd_bus_register(struct device *parent,
- struct nd_bus_descriptor *nd_desc)
+struct nd_bus *__nd_bus_register(struct device *parent,
+ struct nd_bus_descriptor *nd_desc, struct module *module)
{
struct nd_bus *nd_bus = kzalloc(sizeof(*nd_bus), GFP_KERNEL);
int rc;
@@ -148,6 +148,7 @@ struct nd_bus *nd_bus_register(struct device *parent,
return NULL;
}
nd_bus->nd_desc = nd_desc;
+ nd_bus->module = module;
nd_bus->dev.parent = parent;
nd_bus->dev.release = nd_bus_release;
nd_bus->dev.groups = nd_desc->attr_groups;
@@ -171,7 +172,7 @@ struct nd_bus *nd_bus_register(struct device *parent,
put_device(&nd_bus->dev);
return NULL;
}
-EXPORT_SYMBOL_GPL(nd_bus_register);
+EXPORT_SYMBOL_GPL(__nd_bus_register);
static int child_unregister(struct device *dev, void *data)
{
@@ -215,7 +216,12 @@ static __init int libnd_init(void)
rc = nd_dimm_init();
if (rc)
goto err_dimm;
+ rc = nd_region_init();
+ if (rc)
+ goto err_region;
return 0;
+ err_region:
+ nd_dimm_exit();
err_dimm:
nd_bus_exit();
return rc;
@@ -224,6 +230,7 @@ static __init int libnd_init(void)
static __exit void libnd_exit(void)
{
WARN_ON(!list_empty(&nd_bus_list));
+ nd_region_exit();
nd_dimm_exit();
nd_bus_exit();
}
diff --git a/drivers/block/nd/dimm.c b/drivers/block/nd/dimm.c
index a4c8e3ffe97c..6b7d2842509c 100644
--- a/drivers/block/nd/dimm.c
+++ b/drivers/block/nd/dimm.c
@@ -85,7 +85,7 @@ int __init nd_dimm_init(void)
return nd_driver_register(&nd_dimm_driver);
}
-void __exit nd_dimm_exit(void)
+void nd_dimm_exit(void)
{
driver_unregister(&nd_dimm_driver.drv);
}
diff --git a/drivers/block/nd/libnd.h b/drivers/block/nd/libnd.h
index 630a703d1316..8c6f07696f30 100644
--- a/drivers/block/nd/libnd.h
+++ b/drivers/block/nd/libnd.h
@@ -69,8 +69,10 @@ struct nd_region_desc {
};
struct nd_bus;
-struct nd_bus *nd_bus_register(struct device *parent,
- struct nd_bus_descriptor *nfit_desc);
+struct nd_bus *__nd_bus_register(struct device *parent,
+ struct nd_bus_descriptor *nfit_desc, struct module *module);
+#define nd_bus_register(parent, desc) \
+ __nd_bus_register(parent, desc, THIS_MODULE)
void nd_bus_unregister(struct nd_bus *nd_bus);
struct nd_bus *to_nd_bus(struct device *dev);
struct nd_dimm *to_nd_dimm(struct device *dev);
diff --git a/drivers/block/nd/namespace_devs.c b/drivers/block/nd/namespace_devs.c
new file mode 100644
index 000000000000..6861327f4245
--- /dev/null
+++ b/drivers/block/nd/namespace_devs.c
@@ -0,0 +1,111 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/slab.h>
+#include <linux/nd.h>
+#include "nd.h"
+
+static void namespace_io_release(struct device *dev)
+{
+ struct nd_namespace_io *nsio = to_nd_namespace_io(dev);
+
+ kfree(nsio);
+}
+
+static struct device_type namespace_io_device_type = {
+ .name = "nd_namespace_io",
+ .release = namespace_io_release,
+};
+
+static ssize_t type_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_region *nd_region = to_nd_region(dev->parent);
+
+ return sprintf(buf, "%d\n", nd_region_to_namespace_type(nd_region));
+}
+static DEVICE_ATTR_RO(type);
+
+static struct attribute *nd_namespace_attributes[] = {
+ &dev_attr_type.attr,
+ NULL,
+};
+
+static struct attribute_group nd_namespace_attribute_group = {
+ .attrs = nd_namespace_attributes,
+};
+
+static const struct attribute_group *nd_namespace_attribute_groups[] = {
+ &nd_device_attribute_group,
+ &nd_namespace_attribute_group,
+ NULL,
+};
+
+static struct device **create_namespace_io(struct nd_region *nd_region)
+{
+ struct nd_namespace_io *nsio;
+ struct device *dev, **devs;
+ struct resource *res;
+
+ nsio = kzalloc(sizeof(*nsio), GFP_KERNEL);
+ if (!nsio)
+ return NULL;
+
+ devs = kcalloc(2, sizeof(struct device *), GFP_KERNEL);
+ if (!devs) {
+ kfree(nsio);
+ return NULL;
+ }
+
+ dev = &nsio->dev;
+ dev->type = &namespace_io_device_type;
+ res = &nsio->res;
+ res->name = dev_name(&nd_region->dev);
+ res->flags = IORESOURCE_MEM;
+ res->start = nd_region->ndr_start;
+ res->end = res->start + nd_region->ndr_size - 1;
+
+ devs[0] = dev;
+ return devs;
+}
+
+int nd_region_register_namespaces(struct nd_region *nd_region, int *err)
+{
+ struct device **devs = NULL;
+ int i;
+
+ *err = 0;
+ switch (nd_region_to_namespace_type(nd_region)) {
+ case ND_DEVICE_NAMESPACE_IO:
+ devs = create_namespace_io(nd_region);
+ break;
+ default:
+ break;
+ }
+
+ if (!devs)
+ return -ENODEV;
+
+ for (i = 0; devs[i]; i++) {
+ struct device *dev = devs[i];
+
+ dev_set_name(dev, "namespace%d.%d", nd_region->id, i);
+ dev->parent = &nd_region->dev;
+ dev->groups = nd_namespace_attribute_groups;
+ nd_device_register(dev);
+ }
+ kfree(devs);
+
+ return i;
+}
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
index 838b6f958c00..131fc66ce7ab 100644
--- a/drivers/block/nd/nd-private.h
+++ b/drivers/block/nd/nd-private.h
@@ -21,6 +21,7 @@ extern int nd_dimm_major;
struct nd_bus {
struct nd_bus_descriptor *nd_desc;
+ struct module *module;
struct list_head list;
struct device dev;
int id;
@@ -34,16 +35,20 @@ struct nd_dimm {
int id;
};
+bool is_nd_dimm(struct device *dev);
+bool is_nd_blk(struct device *dev);
+bool is_nd_pmem(struct device *dev);
struct nd_bus *walk_to_nd_bus(struct device *nd_dev);
int __init nd_bus_init(void);
void nd_bus_exit(void);
int __init nd_dimm_init(void);
-void __exit nd_dimm_exit(void);
+int __init nd_region_init(void);
+void nd_dimm_exit(void);
+int nd_region_exit(void);
int nd_bus_create_ndctl(struct nd_bus *nd_bus);
void nd_bus_destroy_ndctl(struct nd_bus *nd_bus);
void nd_synchronize(void);
int nd_bus_register_dimms(struct nd_bus *nd_bus);
int nd_bus_register_regions(struct nd_bus *nd_bus);
int nd_match_dimm(struct device *dev, void *data);
-bool is_nd_dimm(struct device *dev);
#endif /* __ND_PRIVATE_H__ */
diff --git a/drivers/block/nd/nd.h b/drivers/block/nd/nd.h
index cae83de12c45..23469513f4c0 100644
--- a/drivers/block/nd/nd.h
+++ b/drivers/block/nd/nd.h
@@ -23,6 +23,11 @@ struct nd_dimm_drvdata {
void *data;
};
+struct nd_region_namespaces {
+ int count;
+ int active;
+};
+
struct nd_region {
struct device dev;
u16 ndr_mappings;
@@ -42,4 +47,7 @@ void nd_device_register(struct device *dev);
void nd_device_unregister(struct device *dev, enum nd_async_mode mode);
int nd_dimm_init_nsarea(struct nd_dimm_drvdata *ndd);
int nd_dimm_init_config_data(struct nd_dimm_drvdata *ndd);
+struct nd_region *to_nd_region(struct device *dev);
+int nd_region_to_namespace_type(struct nd_region *nd_region);
+int nd_region_register_namespaces(struct nd_region *nd_region, int *err);
#endif /* __ND_H__ */
diff --git a/drivers/block/nd/region.c b/drivers/block/nd/region.c
new file mode 100644
index 000000000000..9d1fd45d78a1
--- /dev/null
+++ b/drivers/block/nd/region.c
@@ -0,0 +1,88 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/nd.h>
+#include "nd.h"
+
+static int nd_region_probe(struct device *dev)
+{
+ int err;
+ struct nd_region_namespaces *num_ns;
+ struct nd_region *nd_region = to_nd_region(dev);
+ int rc = nd_region_register_namespaces(nd_region, &err);
+
+ num_ns = devm_kzalloc(dev, sizeof(*num_ns), GFP_KERNEL);
+ if (!num_ns)
+ return -ENOMEM;
+
+ if (rc < 0)
+ return rc;
+
+ num_ns->active = rc;
+ num_ns->count = rc + err;
+ dev_set_drvdata(dev, num_ns);
+
+ if (err == 0)
+ return 0;
+
+ if (rc == err)
+ return -ENODEV;
+
+ /*
+ * Given multiple namespaces per region, we do not want to
+ * disable all the successfully registered peer namespaces upon
+ * a single registration failure. If userspace is missing a
+ * namespace that it expects it can disable/re-enable the region
+ * to retry discovery after correcting the failure.
+ * <regionX>/namespaces returns the current
+ * "<async-registered>/<total>" namespace count.
+ */
+ dev_err(dev, "failed to register %d namespace%s, continuing...\n",
+ err, err == 1 ? "" : "s");
+ return 0;
+}
+
+static int child_unregister(struct device *dev, void *data)
+{
+ nd_device_unregister(dev, ND_SYNC);
+ return 0;
+}
+
+static int nd_region_remove(struct device *dev)
+{
+ device_for_each_child(dev, NULL, child_unregister);
+ return 0;
+}
+
+static struct nd_device_driver nd_region_driver = {
+ .probe = nd_region_probe,
+ .remove = nd_region_remove,
+ .drv = {
+ .name = "nd_region",
+ },
+ .type = ND_DRIVER_REGION_BLK | ND_DRIVER_REGION_PMEM,
+};
+
+int __init nd_region_init(void)
+{
+ return nd_driver_register(&nd_region_driver);
+}
+
+void __exit nd_region_exit(void)
+{
+ driver_unregister(&nd_region_driver.drv);
+}
+
+MODULE_ALIAS_ND_DEVICE(ND_DEVICE_REGION_PMEM);
+MODULE_ALIAS_ND_DEVICE(ND_DEVICE_REGION_BLK);
diff --git a/drivers/block/nd/region_devs.c b/drivers/block/nd/region_devs.c
index 12a5415acfcc..49ebce0c97be 100644
--- a/drivers/block/nd/region_devs.c
+++ b/drivers/block/nd/region_devs.c
@@ -47,11 +47,16 @@ static struct device_type nd_volatile_device_type = {
.release = nd_region_release,
};
-static bool is_nd_pmem(struct device *dev)
+bool is_nd_pmem(struct device *dev)
{
return dev ? dev->type == &nd_pmem_device_type : false;
}
+bool is_nd_blk(struct device *dev)
+{
+ return dev ? dev->type == &nd_blk_device_type : false;
+}
+
struct nd_region *to_nd_region(struct device *dev)
{
struct nd_region *nd_region = container_of(dev, struct nd_region, dev);
@@ -61,6 +66,37 @@ struct nd_region *to_nd_region(struct device *dev)
}
EXPORT_SYMBOL_GPL(to_nd_region);
+/**
+ * nd_region_to_namespace_type() - region to an integer namespace type
+ * @nd_region: region-device to interrogate
+ *
+ * This is the 'nstype' attribute of a region as well, an input to the
+ * MODALIAS for namespace devices, and bit number for a nd_bus to match
+ * namespace devices with namespace drivers.
+ */
+int nd_region_to_namespace_type(struct nd_region *nd_region)
+{
+ if (is_nd_pmem(&nd_region->dev)) {
+ u16 i, alias;
+
+ for (i = 0, alias = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ struct nd_dimm *nd_dimm = nd_mapping->nd_dimm;
+
+ if (nd_dimm->flags & NDD_ALIASING)
+ alias++;
+ }
+ if (alias)
+ return ND_DEVICE_NAMESPACE_PMEM;
+ else
+ return ND_DEVICE_NAMESPACE_IO;
+ } else if (is_nd_blk(&nd_region->dev)) {
+ return ND_DEVICE_NAMESPACE_BLK;
+ }
+
+ return 0;
+}
+
static ssize_t size_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
@@ -88,9 +124,32 @@ static ssize_t mappings_show(struct device *dev,
}
static DEVICE_ATTR_RO(mappings);
+static ssize_t nstype_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_region *nd_region = to_nd_region(dev);
+
+ return sprintf(buf, "%d\n", nd_region_to_namespace_type(nd_region));
+}
+static DEVICE_ATTR_RO(nstype);
+
+static ssize_t init_namespaces_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_region_namespaces *num_ns = dev_get_drvdata(dev);
+
+ if (!num_ns)
+ return -ENXIO;
+
+ return sprintf(buf, "%d/%d\n", num_ns->active, num_ns->count);
+}
+static DEVICE_ATTR_RO(init_namespaces);
+
static struct attribute *nd_region_attributes[] = {
&dev_attr_size.attr,
+ &dev_attr_nstype.attr,
&dev_attr_mappings.attr,
+ &dev_attr_init_namespaces.attr,
NULL,
};
diff --git a/include/linux/nd.h b/include/linux/nd.h
index e074f67e53a3..da70e9962197 100644
--- a/include/linux/nd.h
+++ b/include/linux/nd.h
@@ -26,6 +26,16 @@ static inline struct nd_device_driver *to_nd_device_driver(
struct device_driver *drv)
{
return container_of(drv, struct nd_device_driver, drv);
+};
+
+struct nd_namespace_io {
+ struct device dev;
+ struct resource res;
+};
+
+static inline struct nd_namespace_io *to_nd_namespace_io(struct device *dev)
+{
+ return container_of(dev, struct nd_namespace_io, dev);
}
#define MODULE_ALIAS_ND_DEVICE(type) \
diff --git a/include/uapi/linux/ndctl.h b/include/uapi/linux/ndctl.h
index 1ccd2c633193..5ffa319f3408 100644
--- a/include/uapi/linux/ndctl.h
+++ b/include/uapi/linux/ndctl.h
@@ -177,8 +177,18 @@ static inline const char *nd_dimm_cmd_name(unsigned cmd)
#define ND_DEVICE_DIMM 1 /* nd_dimm: container for "config data" */
+#define ND_DEVICE_REGION_PMEM 2 /* nd_region: (parent of pmem namespaces) */
+#define ND_DEVICE_REGION_BLK 3 /* nd_region: (parent of blk namespaces) */
+#define ND_DEVICE_NAMESPACE_IO 4 /* legacy persistent memory */
+#define ND_DEVICE_NAMESPACE_PMEM 5 /* persistent memory namespace (may alias) */
+#define ND_DEVICE_NAMESPACE_BLK 6 /* block-data-window namespace (may alias) */
enum nd_driver_flags {
ND_DRIVER_DIMM = 1 << ND_DEVICE_DIMM,
+ ND_DRIVER_REGION_PMEM = 1 << ND_DEVICE_REGION_PMEM,
+ ND_DRIVER_REGION_BLK = 1 << ND_DEVICE_REGION_BLK,
+ ND_DRIVER_NAMESPACE_IO = 1 << ND_DEVICE_NAMESPACE_IO,
+ ND_DRIVER_NAMESPACE_PMEM = 1 << ND_DEVICE_NAMESPACE_PMEM,
+ ND_DRIVER_NAMESPACE_BLK = 1 << ND_DEVICE_NAMESPACE_BLK,
};
#endif /* __NDCTL_H__ */
In preparation for the pmem driver attaching to pmem-namespaces emitted
by libnd, convert it to use an ida instead of an always increasing
atomic index. This provides a bit of stability to pmem device names in
the presence of driver re-bind events.
Cc: Christoph Hellwig <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/pmem.c | 22 +++++++++++++++-------
1 file changed, 15 insertions(+), 7 deletions(-)
diff --git a/drivers/block/pmem.c b/drivers/block/pmem.c
index eabf4a8d0085..e3cf9142b172 100644
--- a/drivers/block/pmem.c
+++ b/drivers/block/pmem.c
@@ -34,10 +34,11 @@ struct pmem_device {
phys_addr_t phys_addr;
void *virt_addr;
size_t size;
+ int id;
};
static int pmem_major;
-static atomic_t pmem_index;
+static DEFINE_IDA(pmem_ida);
static void pmem_do_bvec(struct pmem_device *pmem, struct page *page,
unsigned int len, unsigned int off, int rw,
@@ -122,20 +123,26 @@ static struct pmem_device *pmem_alloc(struct device *dev, struct resource *res)
{
struct pmem_device *pmem;
struct gendisk *disk;
- int idx, err;
+ int err;
err = -ENOMEM;
pmem = kzalloc(sizeof(*pmem), GFP_KERNEL);
if (!pmem)
goto out;
+ pmem->id = ida_simple_get(&pmem_ida, 0, 0, GFP_KERNEL);
+ if (pmem->id < 0) {
+ err = pmem->id;
+ goto out_free_dev;
+ }
+
pmem->phys_addr = res->start;
pmem->size = resource_size(res);
err = -EINVAL;
if (!request_mem_region(pmem->phys_addr, pmem->size, "pmem")) {
dev_warn(dev, "could not reserve region [0x%pa:0x%zx]\n", &pmem->phys_addr, pmem->size);
- goto out_free_dev;
+ goto out_free_ida;
}
/*
@@ -159,15 +166,13 @@ static struct pmem_device *pmem_alloc(struct device *dev, struct resource *res)
if (!disk)
goto out_free_queue;
- idx = atomic_inc_return(&pmem_index) - 1;
-
disk->major = pmem_major;
- disk->first_minor = PMEM_MINORS * idx;
+ disk->first_minor = PMEM_MINORS * pmem->id;
disk->fops = &pmem_fops;
disk->private_data = pmem;
disk->queue = pmem->pmem_queue;
disk->flags = GENHD_FL_EXT_DEVT;
- sprintf(disk->disk_name, "pmem%d", idx);
+ sprintf(disk->disk_name, "pmem%d", pmem->id);
disk->driverfs_dev = dev;
set_capacity(disk, pmem->size >> 9);
pmem->pmem_disk = disk;
@@ -182,6 +187,8 @@ out_unmap:
iounmap(pmem->virt_addr);
out_release_region:
release_mem_region(pmem->phys_addr, pmem->size);
+out_free_ida:
+ ida_simple_remove(&pmem_ida, pmem->id);
out_free_dev:
kfree(pmem);
out:
@@ -195,6 +202,7 @@ static void pmem_free(struct pmem_device *pmem)
blk_cleanup_queue(pmem->pmem_queue);
iounmap(pmem->virt_addr);
release_mem_region(pmem->phys_addr, pmem->size);
+ ida_simple_remove(&pmem_ida, pmem->id);
kfree(pmem);
}
nd_pmem attaches to persistent memory regions and namespaces emitted by
the nd subsystem, and, same as the original pmem driver, presents the
system-physical-address range as a block device.
The existing e820-type-12 to pmem setup is converted to a full libnd bus
that emits an nd_namespace_io device.
Cc: Andy Lutomirski <[email protected]>
Cc: Boaz Harrosh <[email protected]>
Cc: H. Peter Anvin <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
arch/x86/kernel/pmem.c | 2 -
drivers/block/Kconfig | 11 -----
drivers/block/Makefile | 1
drivers/block/nd/Kconfig | 27 ++++++++++++
drivers/block/nd/Makefile | 6 +++
drivers/block/nd/e820.c | 100 +++++++++++++++++++++++++++++++++++++++++++++
drivers/block/nd/pmem.c | 47 ++++++++++-----------
7 files changed, 157 insertions(+), 37 deletions(-)
create mode 100644 drivers/block/nd/e820.c
rename drivers/block/{pmem.c => nd/pmem.c} (88%)
diff --git a/arch/x86/kernel/pmem.c b/arch/x86/kernel/pmem.c
index 3420c874ddc5..279328c42f87 100644
--- a/arch/x86/kernel/pmem.c
+++ b/arch/x86/kernel/pmem.c
@@ -13,7 +13,7 @@ static __init void register_pmem_device(struct resource *res)
struct platform_device *pdev;
int error;
- pdev = platform_device_alloc("pmem", PLATFORM_DEVID_AUTO);
+ pdev = platform_device_alloc("e820_pmem", PLATFORM_DEVID_AUTO);
if (!pdev)
return;
diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig
index dfe40e5ca9bd..1cef4ffb16c5 100644
--- a/drivers/block/Kconfig
+++ b/drivers/block/Kconfig
@@ -406,17 +406,6 @@ config BLK_DEV_RAM_DAX
and will prevent RAM block device backing store memory from being
allocated from highmem (only a problem for highmem systems).
-config BLK_DEV_PMEM
- tristate "Persistent memory block device support"
- help
- Saying Y here will allow you to use a contiguous range of reserved
- memory as one or more persistent block devices.
-
- To compile this driver as a module, choose M here: the module will be
- called 'pmem'.
-
- If unsure, say N.
-
config CDROM_PKTCDVD
tristate "Packet writing on CD/DVD media"
depends on !UML
diff --git a/drivers/block/Makefile b/drivers/block/Makefile
index 07a6acecf4d8..964d8eb2c16f 100644
--- a/drivers/block/Makefile
+++ b/drivers/block/Makefile
@@ -14,7 +14,6 @@ obj-$(CONFIG_PS3_VRAM) += ps3vram.o
obj-$(CONFIG_ATARI_FLOPPY) += ataflop.o
obj-$(CONFIG_AMIGA_Z2RAM) += z2ram.o
obj-$(CONFIG_BLK_DEV_RAM) += brd.o
-obj-$(CONFIG_BLK_DEV_PMEM) += pmem.o
obj-$(CONFIG_BLK_DEV_LOOP) += loop.o
obj-$(CONFIG_BLK_CPQ_DA) += cpqarray.o
obj-$(CONFIG_BLK_CPQ_CISS_DA) += cciss.o
diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
index d2d84451e82c..c5eaf195734d 100644
--- a/drivers/block/nd/Kconfig
+++ b/drivers/block/nd/Kconfig
@@ -68,4 +68,31 @@ config NFIT_TEST
Say N unless you are doing development of the 'nd' subsystem.
+config ND_E820
+ tristate "E820: Support the E820-type-12 PMEM convention"
+ depends on X86_PMEM_LEGACY
+ default m if X86_PMEM_LEGACY
+ select LIBND
+ help
+ Prior to ACPI 6 some platforms advertised peristent memory
+ via type-12 e820 memory ranges. Create a libnd bus and
+ attach an instance of the pmem driver to these ranges.
+
+config BLK_DEV_PMEM
+ tristate "PMEM: Persistent memory block device support"
+ depends on LIBND
+ default LIBND
+ help
+ Memory ranges for PMEM are described by either an NFIT
+ (NVDIMM Firmware Interface Table, see CONFIG_NFIT_ACPI), a
+ non-standard OEM-specific E820 memory type (type-12, see
+ CONFIG_X86_PMEM_LEGACY), or it is manually specified by the
+ 'memmap=nn[KMG]!ss[KMG]' kernel command line (see
+ Documentation/kernel-parameters.txt). This driver converts
+ these persistent memory ranges into block devices that are
+ capable of DAX (direct-access) file system mappings. See
+ Documentation/blockdev/nd.txt for more details.
+
+ Say Y if you want to use a NVDIMM described by NFIT
+
endif
diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
index 0fb0891e1817..ebb212af9f15 100644
--- a/drivers/block/nd/Makefile
+++ b/drivers/block/nd/Makefile
@@ -14,10 +14,16 @@ endif
obj-$(CONFIG_LIBND) += libnd.o
obj-$(CONFIG_ND_ACPI) += nd_acpi.o
+obj-$(CONFIG_ND_E820) += nd_e820.o
obj-$(CONFIG_NFIT_TEST) += test/
+obj-$(CONFIG_BLK_DEV_PMEM) += nd_pmem.o
nd_acpi-y := acpi.o
+nd_e820-y := e820.o
+
+nd_pmem-y := pmem.o
+
libnd-y := core.o
libnd-y += bus.o
libnd-y += dimm_devs.o
diff --git a/drivers/block/nd/e820.c b/drivers/block/nd/e820.c
new file mode 100644
index 000000000000..f4db8c54248e
--- /dev/null
+++ b/drivers/block/nd/e820.c
@@ -0,0 +1,100 @@
+/*
+ * libnd e820 support
+ *
+ * Copyright (c) 2014-2015, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/module.h>
+#include "libnd.h"
+
+static const struct attribute_group *nd_e820_attribute_groups[] = {
+ &nd_bus_attribute_group,
+ NULL,
+};
+
+static const struct attribute_group *nd_e820_region_attribute_groups[] = {
+ &nd_region_attribute_group,
+ &nd_device_attribute_group,
+ NULL,
+};
+
+static int nd_e820_probe(struct platform_device *pdev)
+{
+ struct nd_bus_descriptor *nd_desc;
+ struct nd_region_desc ndr_desc;
+ struct nd_bus *nd_bus;
+ struct resource *res;
+
+ if (WARN_ON(pdev->num_resources > 1))
+ return -ENXIO;
+
+ res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+ if (!res)
+ return -ENXIO;
+
+ nd_desc = devm_kzalloc(&pdev->dev, sizeof(*nd_desc), GFP_KERNEL);
+ if (!nd_desc)
+ return -ENOMEM;
+
+ nd_desc->attr_groups = nd_e820_attribute_groups;
+ nd_desc->provider_name = "e820";
+ nd_bus = nd_bus_register(&pdev->dev, nd_desc);
+ if (!nd_bus)
+ return -ENXIO;
+
+ memset(&ndr_desc, 0, sizeof(ndr_desc));
+ ndr_desc.res = res;
+ ndr_desc.attr_groups = nd_e820_region_attribute_groups;
+ if (!nd_pmem_region_create(nd_bus, &ndr_desc)) {
+ nd_bus_unregister(nd_bus);
+ return -ENXIO;
+ }
+
+ platform_set_drvdata(pdev, nd_bus);
+
+ return 0;
+}
+
+static int nd_e820_remove(struct platform_device *pdev)
+{
+ struct nd_bus *nd_bus = platform_get_drvdata(pdev);
+
+ nd_bus_unregister(nd_bus);
+
+ return 0;
+}
+
+static struct platform_driver nd_e820_driver = {
+ .probe = nd_e820_probe,
+ .remove = nd_e820_remove,
+ .driver = {
+ .owner = THIS_MODULE,
+ .name = "e820_pmem",
+ },
+};
+
+MODULE_ALIAS("platform:e820_pmem*");
+
+static int __init nd_e820_init(void)
+{
+ return platform_driver_register(&nd_e820_driver);
+}
+module_init(nd_e820_init);
+
+static void nd_e820_exit(void)
+{
+ platform_driver_unregister(&nd_e820_driver);
+}
+module_exit(nd_e820_exit);
+
+MODULE_AUTHOR("Intel Corporation");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/block/pmem.c b/drivers/block/nd/pmem.c
similarity index 88%
rename from drivers/block/pmem.c
rename to drivers/block/nd/pmem.c
index e3cf9142b172..c9dd8d7eca3a 100644
--- a/drivers/block/pmem.c
+++ b/drivers/block/nd/pmem.c
@@ -1,7 +1,7 @@
/*
* Persistent Memory Driver
*
- * Copyright (c) 2014, Intel Corporation.
+ * Copyright (c) 2014-2015, Intel Corporation.
* Copyright (c) 2015, Christoph Hellwig <[email protected]>.
* Copyright (c) 2015, Boaz Harrosh <[email protected]>.
*
@@ -23,6 +23,7 @@
#include <linux/module.h>
#include <linux/moduleparam.h>
#include <linux/slab.h>
+#include <linux/nd.h>
#define PMEM_MINORS 16
@@ -206,42 +207,37 @@ static void pmem_free(struct pmem_device *pmem)
kfree(pmem);
}
-static int pmem_probe(struct platform_device *pdev)
+static int nd_pmem_probe(struct device *dev)
{
+ struct nd_namespace_io *nsio = to_nd_namespace_io(dev);
struct pmem_device *pmem;
- struct resource *res;
- if (WARN_ON(pdev->num_resources > 1))
- return -ENXIO;
-
- res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
- if (!res)
- return -ENXIO;
-
- pmem = pmem_alloc(&pdev->dev, res);
+ pmem = pmem_alloc(dev, &nsio->res);
if (IS_ERR(pmem))
return PTR_ERR(pmem);
- platform_set_drvdata(pdev, pmem);
+ dev_set_drvdata(dev, pmem);
return 0;
}
-static int pmem_remove(struct platform_device *pdev)
+static int nd_pmem_remove(struct device *dev)
{
- struct pmem_device *pmem = platform_get_drvdata(pdev);
+ struct pmem_device *pmem = dev_get_drvdata(dev);
pmem_free(pmem);
return 0;
}
-static struct platform_driver pmem_driver = {
- .probe = pmem_probe,
- .remove = pmem_remove,
- .driver = {
- .owner = THIS_MODULE,
- .name = "pmem",
+MODULE_ALIAS("pmem");
+MODULE_ALIAS_ND_DEVICE(ND_DEVICE_NAMESPACE_IO);
+static struct nd_device_driver nd_pmem_driver = {
+ .probe = nd_pmem_probe,
+ .remove = nd_pmem_remove,
+ .drv = {
+ .name = "pmem",
},
+ .type = ND_DRIVER_NAMESPACE_IO,
};
static int __init pmem_init(void)
@@ -252,16 +248,19 @@ static int __init pmem_init(void)
if (pmem_major < 0)
return pmem_major;
- error = platform_driver_register(&pmem_driver);
- if (error)
+ error = nd_driver_register(&nd_pmem_driver);
+ if (error) {
unregister_blkdev(pmem_major, "pmem");
- return error;
+ return error;
+ }
+
+ return 0;
}
module_init(pmem_init);
static void pmem_exit(void)
{
- platform_driver_unregister(&pmem_driver);
+ driver_unregister(&nd_pmem_driver.drv);
unregister_blkdev(pmem_major, "pmem");
}
module_exit(pmem_exit);
On platforms that have firmware support for reading/writing per-dimm
label space, a portion of the dimm may be accessible via an interleave
set PMEM mapping in addition to the dimm's BLK (block-data-window
aperture(s)) interface. A label, stored in a "configuration data
region" on the dimm, disambiguates which dimm addresses are accessed
through which exclusive interface.
Add infrastructure that allows the kernel to block modifications to a
label in the set while any member dimm is active. Note that this is
meant only for enforcing "no modifications of active labels" via the
coarse ioctl command. Adding/deleting namespaces from an active
interleave set will only be possible via sysfs.
Another aspect of tracking interleave sets is tracking their integrity
when DIMMs in a set are physically re-ordered. For this purpose we
generate an "interleave-set cookie" that can be recorded in a label and
validated against the current configuration. It is the bus provider
implementation's responsibility to calculate the interleave set cookie
and attach it to a given region.
Cc: Neil Brown <[email protected]>
Cc: <[email protected]>
Cc: Greg KH <[email protected]>
Cc: Robert Moore <[email protected]>
Cc: Rafael J. Wysocki <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/acpi.c | 90 ++++++++++++++++++++++++++++++++++++++++
drivers/block/nd/bus.c | 41 ++++++++++++++++++
drivers/block/nd/core.c | 47 +++++++++++++++++++++
drivers/block/nd/dimm_devs.c | 19 ++++++++
drivers/block/nd/libnd.h | 6 +++
drivers/block/nd/nd-private.h | 11 ++++-
drivers/block/nd/nd.h | 4 ++
drivers/block/nd/region_devs.c | 85 ++++++++++++++++++++++++++++++++++++++
8 files changed, 299 insertions(+), 4 deletions(-)
diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
index c3dda74f73d7..d34cefe38e2f 100644
--- a/drivers/block/nd/acpi.c
+++ b/drivers/block/nd/acpi.c
@@ -15,6 +15,7 @@
#include <linux/ndctl.h>
#include <linux/list.h>
#include <linux/acpi.h>
+#include <linux/sort.h>
#include "acpi_nfit.h"
#include "libnd.h"
@@ -779,6 +780,90 @@ static const struct attribute_group *nd_acpi_region_attribute_groups[] = {
NULL,
};
+/* enough info to uniquely specify an interleave set */
+struct nfit_set_info {
+ struct nfit_set_info_map {
+ u64 region_spa_offset;
+ u32 serial_number;
+ u32 pad;
+ } mapping[0];
+};
+
+static size_t sizeof_nfit_set_info(int num_mappings)
+{
+ return sizeof(struct nfit_set_info)
+ + num_mappings * sizeof(struct nfit_set_info_map);
+}
+
+static int cmp_map(const void *m0, const void *m1)
+{
+ const struct nfit_set_info_map *map0 = m0;
+ const struct nfit_set_info_map *map1 = m1;
+
+ return memcmp(&map0->region_spa_offset, &map1->region_spa_offset,
+ sizeof(u64));
+}
+
+/* Retrieve the nth entry referencing this spa */
+static struct acpi_nfit_memdev *memdev_from_spa(
+ struct acpi_nfit_desc *acpi_desc, u16 spa_index, int n)
+{
+ struct nfit_memdev *nfit_memdev;
+
+ list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list)
+ if (nfit_memdev->memdev->spa_index == spa_index)
+ if (n-- == 0)
+ return nfit_memdev->memdev;
+ return NULL;
+}
+
+static int nd_acpi_init_interleave_set(struct acpi_nfit_desc *acpi_desc,
+ struct nd_region_desc *ndr_desc, struct acpi_nfit_spa *spa)
+{
+ u16 num_mappings = ndr_desc->num_mappings;
+ int i, spa_type = nfit_spa_type(spa);
+ struct device *dev = acpi_desc->dev;
+ struct nd_interleave_set *nd_set;
+ struct nfit_set_info *info;
+
+ if (spa_type == NFIT_SPA_PM || spa_type == NFIT_SPA_VOLATILE)
+ /* pass */;
+ else
+ return 0;
+
+ nd_set = devm_kzalloc(dev, sizeof(*nd_set), GFP_KERNEL);
+ if (!nd_set)
+ return -ENOMEM;
+
+ info = devm_kzalloc(dev, sizeof_nfit_set_info(num_mappings), GFP_KERNEL);
+ if (!info)
+ return -ENOMEM;
+ for (i = 0; i < num_mappings; i++) {
+ struct nd_mapping *nd_mapping = &ndr_desc->nd_mapping[i];
+ struct nfit_set_info_map *map = &info->mapping[i];
+ struct nd_dimm *nd_dimm = nd_mapping->nd_dimm;
+ struct nfit_mem *nfit_mem = nd_dimm_provider_data(nd_dimm);
+ struct acpi_nfit_memdev *memdev = memdev_from_spa(acpi_desc,
+ spa->spa_index, i);
+
+ if (!memdev || !nfit_mem->dcr) {
+ dev_err(dev, "%s: failed to find DCR\n", __func__);
+ return -ENODEV;
+ }
+
+ map->region_spa_offset = memdev->region_spa_offset;
+ map->serial_number = nfit_mem->dcr->serial_number;
+ }
+
+ sort(&info->mapping[0], num_mappings, sizeof(struct nfit_set_info_map),
+ cmp_map, NULL);
+ nd_set->cookie = nd_fletcher64(info, sizeof_nfit_set_info(num_mappings));
+ ndr_desc->nd_set = nd_set;
+ devm_kfree(dev, info);
+
+ return 0;
+}
+
static int nd_acpi_register_region(struct acpi_nfit_desc *acpi_desc,
struct nfit_spa *nfit_spa)
{
@@ -786,7 +871,7 @@ static int nd_acpi_register_region(struct acpi_nfit_desc *acpi_desc,
struct acpi_nfit_spa *spa = nfit_spa->spa;
struct nfit_memdev *nfit_memdev;
struct nd_region_desc ndr_desc;
- int spa_type, count = 0;
+ int spa_type, count = 0, rc;
struct resource res;
u16 spa_index;
@@ -852,6 +937,9 @@ static int nd_acpi_register_region(struct acpi_nfit_desc *acpi_desc,
ndr_desc.nd_mapping = nd_mappings;
ndr_desc.num_mappings = count;
+ rc = nd_acpi_init_interleave_set(acpi_desc, &ndr_desc, spa);
+ if (rc)
+ return rc;
if (spa_type == NFIT_SPA_PM) {
if (!nd_pmem_region_create(acpi_desc->nd_bus, &ndr_desc))
return -ENOMEM;
diff --git a/drivers/block/nd/bus.c b/drivers/block/nd/bus.c
index 46568d182559..8afb8d4a7e81 100644
--- a/drivers/block/nd/bus.c
+++ b/drivers/block/nd/bus.c
@@ -78,7 +78,10 @@ static int nd_bus_probe(struct device *dev)
if (!try_module_get(provider))
return -ENXIO;
+ nd_region_probe_start(nd_bus, dev);
rc = nd_drv->probe(dev);
+ nd_region_probe_end(nd_bus, dev, rc);
+
dev_dbg(&nd_bus->dev, "%s.probe(%s) = %d\n", dev->driver->name,
dev_name(dev), rc);
if (rc != 0)
@@ -94,6 +97,8 @@ static int nd_bus_remove(struct device *dev)
int rc;
rc = nd_drv->remove(dev);
+ nd_region_notify_remove(nd_bus, dev, rc);
+
dev_dbg(&nd_bus->dev, "%s.remove(%s) = %d\n", dev->driver->name,
dev_name(dev), rc);
module_put(provider);
@@ -359,6 +364,33 @@ u32 nd_cmd_out_size(struct nd_dimm *nd_dimm, int cmd,
}
EXPORT_SYMBOL_GPL(nd_cmd_out_size);
+static void wait_nd_bus_probe_idle(struct nd_bus *nd_bus)
+{
+ do {
+ if (nd_bus->probe_active == 0)
+ break;
+ nd_bus_unlock(&nd_bus->dev);
+ wait_event(nd_bus->probe_wait, nd_bus->probe_active == 0);
+ nd_bus_lock(&nd_bus->dev);
+ } while (true);
+}
+
+/* set_config requires an idle interleave set */
+static int nd_cmd_clear_to_send(struct nd_dimm *nd_dimm, unsigned int cmd)
+{
+ struct nd_bus *nd_bus;
+
+ if (!nd_dimm || cmd != ND_CMD_SET_CONFIG_DATA)
+ return 0;
+
+ nd_bus = walk_to_nd_bus(&nd_dimm->dev);
+ wait_nd_bus_probe_idle(nd_bus);
+
+ if (atomic_read(&nd_dimm->busy))
+ return -EBUSY;
+ return 0;
+}
+
static int __nd_ioctl(struct nd_bus *nd_bus, struct nd_dimm *nd_dimm,
int read_only, unsigned int ioctl_cmd, unsigned long arg)
{
@@ -469,11 +501,18 @@ static int __nd_ioctl(struct nd_bus *nd_bus, struct nd_dimm *nd_dimm,
goto out;
}
+ nd_bus_lock(&nd_bus->dev);
+ rc = nd_cmd_clear_to_send(nd_dimm, cmd);
+ if (rc)
+ goto out_unlock;
+
rc = nd_desc->ndctl(nd_desc, nd_dimm, cmd, buf, buf_len);
if (rc < 0)
- goto out;
+ goto out_unlock;
if (copy_to_user(p, buf, buf_len))
rc = -EFAULT;
+ out_unlock:
+ nd_bus_unlock(&nd_bus->dev);
out:
vfree(buf);
return rc;
diff --git a/drivers/block/nd/core.c b/drivers/block/nd/core.c
index 646e424ae36c..603970d0ef3a 100644
--- a/drivers/block/nd/core.c
+++ b/drivers/block/nd/core.c
@@ -24,6 +24,51 @@ LIST_HEAD(nd_bus_list);
DEFINE_MUTEX(nd_bus_list_mutex);
static DEFINE_IDA(nd_ida);
+void nd_bus_lock(struct device *dev)
+{
+ struct nd_bus *nd_bus = walk_to_nd_bus(dev);
+
+ if (!nd_bus)
+ return;
+ mutex_lock(&nd_bus->reconfig_mutex);
+}
+EXPORT_SYMBOL(nd_bus_lock);
+
+void nd_bus_unlock(struct device *dev)
+{
+ struct nd_bus *nd_bus = walk_to_nd_bus(dev);
+
+ if (!nd_bus)
+ return;
+ mutex_unlock(&nd_bus->reconfig_mutex);
+}
+EXPORT_SYMBOL(nd_bus_unlock);
+
+bool is_nd_bus_locked(struct device *dev)
+{
+ struct nd_bus *nd_bus = walk_to_nd_bus(dev);
+
+ if (!nd_bus)
+ return false;
+ return mutex_is_locked(&nd_bus->reconfig_mutex);
+}
+EXPORT_SYMBOL(is_nd_bus_locked);
+
+u64 nd_fletcher64(void __iomem *addr, size_t len)
+{
+ u32 lo32 = 0;
+ u64 hi32 = 0;
+ int i;
+
+ for (i = 0; i < len; i += 4) {
+ lo32 = readl(addr + i);
+ hi32 += lo32;
+ }
+
+ return hi32 << 32 | lo32;
+}
+EXPORT_SYMBOL_GPL(nd_fletcher64);
+
static void nd_bus_release(struct device *dev)
{
struct nd_bus *nd_bus = container_of(dev, struct nd_bus, dev);
@@ -142,7 +187,9 @@ struct nd_bus *__nd_bus_register(struct device *parent,
if (!nd_bus)
return NULL;
INIT_LIST_HEAD(&nd_bus->list);
+ init_waitqueue_head(&nd_bus->probe_wait);
nd_bus->id = ida_simple_get(&nd_ida, 0, 0, GFP_KERNEL);
+ mutex_init(&nd_bus->reconfig_mutex);
if (nd_bus->id < 0) {
kfree(nd_bus);
return NULL;
diff --git a/drivers/block/nd/dimm_devs.c b/drivers/block/nd/dimm_devs.c
index 33b6d5336096..8981adc59ba4 100644
--- a/drivers/block/nd/dimm_devs.c
+++ b/drivers/block/nd/dimm_devs.c
@@ -185,7 +185,24 @@ static ssize_t commands_show(struct device *dev,
}
static DEVICE_ATTR_RO(commands);
+static ssize_t state_show(struct device *dev, struct device_attribute *attr,
+ char *buf)
+{
+ struct nd_dimm *nd_dimm = to_nd_dimm(dev);
+
+ /*
+ * The state may be in the process of changing, userspace should
+ * quiesce probing if it wants a static answer
+ */
+ nd_bus_lock(dev);
+ nd_bus_unlock(dev);
+ return sprintf(buf, "%s\n", atomic_read(&nd_dimm->busy)
+ ? "active" : "idle");
+}
+static DEVICE_ATTR_RO(state);
+
static struct attribute *nd_dimm_attributes[] = {
+ &dev_attr_state.attr,
&dev_attr_commands.attr,
NULL,
};
@@ -213,7 +230,7 @@ struct nd_dimm *nd_dimm_create(struct nd_bus *nd_bus, void *provider_data,
nd_dimm->provider_data = provider_data;
nd_dimm->flags = flags;
nd_dimm->dsm_mask = dsm_mask;
-
+ atomic_set(&nd_dimm->busy, 0);
dev = &nd_dimm->dev;
dev_set_name(dev, "nmem%d", nd_dimm->id);
dev->parent = &nd_bus->dev;
diff --git a/drivers/block/nd/libnd.h b/drivers/block/nd/libnd.h
index 8c6f07696f30..deb29eff5b61 100644
--- a/drivers/block/nd/libnd.h
+++ b/drivers/block/nd/libnd.h
@@ -60,11 +60,16 @@ struct nd_cmd_desc {
int out_sizes[ND_CMD_MAX_ELEM];
};
+struct nd_interleave_set {
+ u64 cookie;
+};
+
struct nd_region_desc {
struct resource *res;
struct nd_mapping *nd_mapping;
u16 num_mappings;
const struct attribute_group **attr_groups;
+ struct nd_interleave_set *nd_set;
void *provider_data;
};
@@ -98,4 +103,5 @@ struct nd_region *nd_blk_region_create(struct nd_bus *nd_bus,
struct nd_region_desc *ndr_desc);
struct nd_region *nd_volatile_region_create(struct nd_bus *nd_bus,
struct nd_region_desc *ndr_desc);
+u64 nd_fletcher64(void __iomem *addr, size_t len);
#endif /* __LIBND_H__ */
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
index 131fc66ce7ab..5d8249be3415 100644
--- a/drivers/block/nd/nd-private.h
+++ b/drivers/block/nd/nd-private.h
@@ -13,6 +13,8 @@
#ifndef __ND_PRIVATE_H__
#define __ND_PRIVATE_H__
#include <linux/device.h>
+#include <linux/sizes.h>
+#include <linux/mutex.h>
#include "libnd.h"
extern struct list_head nd_bus_list;
@@ -21,10 +23,12 @@ extern int nd_dimm_major;
struct nd_bus {
struct nd_bus_descriptor *nd_desc;
+ wait_queue_head_t probe_wait;
struct module *module;
struct list_head list;
struct device dev;
- int id;
+ int id, probe_active;
+ struct mutex reconfig_mutex;
};
struct nd_dimm {
@@ -32,6 +36,7 @@ struct nd_dimm {
void *provider_data;
unsigned long *dsm_mask;
struct device dev;
+ atomic_t busy;
int id;
};
@@ -45,10 +50,14 @@ int __init nd_dimm_init(void);
int __init nd_region_init(void);
void nd_dimm_exit(void);
int nd_region_exit(void);
+void nd_region_probe_start(struct nd_bus *nd_bus, struct device *dev);
+void nd_region_probe_end(struct nd_bus *nd_bus, struct device *dev, int rc);
+void nd_region_notify_remove(struct nd_bus *nd_bus, struct device *dev, int rc);
int nd_bus_create_ndctl(struct nd_bus *nd_bus);
void nd_bus_destroy_ndctl(struct nd_bus *nd_bus);
void nd_synchronize(void);
int nd_bus_register_dimms(struct nd_bus *nd_bus);
int nd_bus_register_regions(struct nd_bus *nd_bus);
+int nd_bus_init_interleave_sets(struct nd_bus *nd_bus);
int nd_match_dimm(struct device *dev, void *data);
#endif /* __ND_PRIVATE_H__ */
diff --git a/drivers/block/nd/nd.h b/drivers/block/nd/nd.h
index 23469513f4c0..c69707dbd272 100644
--- a/drivers/block/nd/nd.h
+++ b/drivers/block/nd/nd.h
@@ -35,6 +35,7 @@ struct nd_region {
u64 ndr_start;
int id;
void *provider_data;
+ struct nd_interleave_set *nd_set;
struct nd_mapping mapping[0];
};
@@ -50,4 +51,7 @@ int nd_dimm_init_config_data(struct nd_dimm_drvdata *ndd);
struct nd_region *to_nd_region(struct device *dev);
int nd_region_to_namespace_type(struct nd_region *nd_region);
int nd_region_register_namespaces(struct nd_region *nd_region, int *err);
+void nd_bus_lock(struct device *dev);
+void nd_bus_unlock(struct device *dev);
+bool is_nd_bus_locked(struct device *dev);
#endif /* __ND_H__ */
diff --git a/drivers/block/nd/region_devs.c b/drivers/block/nd/region_devs.c
index 49ebce0c97be..c1cea42d0473 100644
--- a/drivers/block/nd/region_devs.c
+++ b/drivers/block/nd/region_devs.c
@@ -10,7 +10,10 @@
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*/
+#include <linux/scatterlist.h>
+#include <linux/sched.h>
#include <linux/slab.h>
+#include <linux/sort.h>
#include <linux/io.h>
#include "nd-private.h"
#include "nd.h"
@@ -133,6 +136,21 @@ static ssize_t nstype_show(struct device *dev,
}
static DEVICE_ATTR_RO(nstype);
+static ssize_t set_cookie_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_region *nd_region = to_nd_region(dev);
+ struct nd_interleave_set *nd_set = nd_region->nd_set;
+
+ if (is_nd_pmem(dev) && nd_set)
+ /* pass, should be precluded by nd_region_visible */;
+ else
+ return -ENXIO;
+
+ return sprintf(buf, "%#llx\n", nd_set->cookie);
+}
+static DEVICE_ATTR_RO(set_cookie);
+
static ssize_t init_namespaces_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
@@ -149,15 +167,81 @@ static struct attribute *nd_region_attributes[] = {
&dev_attr_size.attr,
&dev_attr_nstype.attr,
&dev_attr_mappings.attr,
+ &dev_attr_set_cookie.attr,
&dev_attr_init_namespaces.attr,
NULL,
};
+static umode_t nd_region_visible(struct kobject *kobj, struct attribute *a, int n)
+{
+ struct device *dev = container_of(kobj, typeof(*dev), kobj);
+ struct nd_region *nd_region = to_nd_region(dev);
+ struct nd_interleave_set *nd_set = nd_region->nd_set;
+
+ if (a != &dev_attr_set_cookie.attr)
+ return a->mode;
+
+ if (is_nd_pmem(dev) && nd_set)
+ return a->mode;
+
+ return 0;
+}
+
struct attribute_group nd_region_attribute_group = {
.attrs = nd_region_attributes,
+ .is_visible = nd_region_visible,
};
EXPORT_SYMBOL_GPL(nd_region_attribute_group);
+/*
+ * Upon successful probe/remove, take/release a reference on the
+ * associated interleave set (if present)
+ */
+static void nd_region_notify_driver_action(struct nd_bus *nd_bus,
+ struct device *dev, int rc, bool probe)
+{
+ if (rc)
+ return;
+
+ if (is_nd_pmem(dev) || is_nd_blk(dev)) {
+ struct nd_region *nd_region = to_nd_region(dev);
+ int i;
+
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ struct nd_dimm *nd_dimm = nd_mapping->nd_dimm;
+
+ if (probe)
+ atomic_inc(&nd_dimm->busy);
+ else
+ atomic_dec(&nd_dimm->busy);
+ }
+ }
+}
+
+void nd_region_probe_start(struct nd_bus *nd_bus, struct device *dev)
+{
+ nd_bus_lock(&nd_bus->dev);
+ nd_bus->probe_active++;
+ nd_bus_unlock(&nd_bus->dev);
+}
+
+void nd_region_probe_end(struct nd_bus *nd_bus, struct device *dev, int rc)
+{
+ nd_bus_lock(&nd_bus->dev);
+ nd_region_notify_driver_action(nd_bus, dev, rc, true);
+ if (--nd_bus->probe_active == 0)
+ wake_up(&nd_bus->probe_wait);
+ nd_bus_unlock(&nd_bus->dev);
+}
+
+void nd_region_notify_remove(struct nd_bus *nd_bus, struct device *dev, int rc)
+{
+ nd_bus_lock(dev);
+ nd_region_notify_driver_action(nd_bus, dev, rc, false);
+ nd_bus_unlock(dev);
+}
+
static ssize_t mappingN(struct device *dev, char *buf, int n)
{
struct nd_region *nd_region = to_nd_region(dev);
@@ -317,6 +401,7 @@ static noinline struct nd_region *nd_region_create(struct nd_bus *nd_bus,
}
nd_region->ndr_mappings = ndr_desc->num_mappings;
nd_region->provider_data = ndr_desc->provider_data;
+ nd_region->nd_set = ndr_desc->nd_set;
dev = &nd_region->dev;
dev_set_name(dev, "region%d", nd_region->id);
dev->parent = &nd_bus->dev;
On media label format consists of two index blocks followed by an array
of labels. None of these structures are ever updated in place. A
sequence number tracks the current active index and the next one to
write, while labels are written to free slots.
+------------+
| |
| nsindex0 |
| |
+------------+
| |
| nsindex1 |
| |
+------------+
| label0 |
+------------+
| label1 |
+------------+
| |
....nslot...
| |
+------------+
| labelN |
+------------+
After reading valid labels, store the dpa ranges they claim into
per-dimm resource trees.
Cc: Neil Brown <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/Makefile | 1
drivers/block/nd/dimm.c | 26 +++-
drivers/block/nd/dimm_devs.c | 6 +
drivers/block/nd/label.c | 291 ++++++++++++++++++++++++++++++++++++++++++
drivers/block/nd/label.h | 129 +++++++++++++++++++
drivers/block/nd/nd.h | 45 ++++++
include/uapi/linux/ndctl.h | 1
7 files changed, 495 insertions(+), 4 deletions(-)
create mode 100644 drivers/block/nd/label.c
create mode 100644 drivers/block/nd/label.h
diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
index ebb212af9f15..d588f691163c 100644
--- a/drivers/block/nd/Makefile
+++ b/drivers/block/nd/Makefile
@@ -31,3 +31,4 @@ libnd-y += dimm.o
libnd-y += region_devs.o
libnd-y += region.o
libnd-y += namespace_devs.o
+libnd-y += label.o
diff --git a/drivers/block/nd/dimm.c b/drivers/block/nd/dimm.c
index 6b7d2842509c..5477176c5de0 100644
--- a/drivers/block/nd/dimm.c
+++ b/drivers/block/nd/dimm.c
@@ -18,6 +18,7 @@
#include <linux/slab.h>
#include <linux/mm.h>
#include <linux/nd.h>
+#include "label.h"
#include "nd.h"
static void free_data(struct nd_dimm_drvdata *ndd)
@@ -42,7 +43,12 @@ static int nd_dimm_probe(struct device *dev)
return -ENOMEM;
dev_set_drvdata(dev, ndd);
- ndd->dev = dev;
+ ndd->dpa.name = dev_name(dev);
+ ndd->ns_current = -1;
+ ndd->ns_next = -1;
+ ndd->dpa.start = 0;
+ ndd->dpa.end = -1;
+ ndd->dev = dev;
rc = nd_dimm_init_nsarea(ndd);
if (rc)
@@ -54,18 +60,34 @@ static int nd_dimm_probe(struct device *dev)
dev_dbg(dev, "config data size: %d\n", ndd->nsarea.config_size);
+ nd_bus_lock(dev);
+ ndd->ns_current = nd_label_validate(ndd);
+ ndd->ns_next = nd_label_next_nsindex(ndd->ns_current);
+ nd_label_copy(ndd, to_next_namespace_index(ndd),
+ to_current_namespace_index(ndd));
+ rc = nd_label_reserve_dpa(ndd);
+ nd_bus_unlock(dev);
+
+ if (rc)
+ goto err;
+
return 0;
err:
free_data(ndd);
return rc;
-
}
static int nd_dimm_remove(struct device *dev)
{
struct nd_dimm_drvdata *ndd = dev_get_drvdata(dev);
+ struct resource *res, *_r;
+ nd_bus_lock(dev);
+ dev_set_drvdata(dev, NULL);
+ for_each_dpa_resource_safe(ndd, res, _r)
+ __release_region(&ndd->dpa, res->start, resource_size(res));
+ nd_bus_unlock(dev);
free_data(ndd);
return 0;
diff --git a/drivers/block/nd/dimm_devs.c b/drivers/block/nd/dimm_devs.c
index 8981adc59ba4..3fbd0d0502eb 100644
--- a/drivers/block/nd/dimm_devs.c
+++ b/drivers/block/nd/dimm_devs.c
@@ -92,8 +92,12 @@ int nd_dimm_init_config_data(struct nd_dimm_drvdata *ndd)
if (ndd->data)
return 0;
- if (ndd->nsarea.status || ndd->nsarea.max_xfer == 0)
+ if (ndd->nsarea.status || ndd->nsarea.max_xfer == 0
+ || ndd->nsarea.config_size < ND_LABEL_MIN_SIZE) {
+ dev_dbg(ndd->dev, "failed to init config data area: (%d:%d)\n",
+ ndd->nsarea.max_xfer, ndd->nsarea.config_size);
return -ENXIO;
+ }
ndd->data = kmalloc(ndd->nsarea.config_size, GFP_KERNEL);
if (!ndd->data)
diff --git a/drivers/block/nd/label.c b/drivers/block/nd/label.c
new file mode 100644
index 000000000000..e791ea8bbdde
--- /dev/null
+++ b/drivers/block/nd/label.c
@@ -0,0 +1,291 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#include <linux/device.h>
+#include <linux/ndctl.h>
+#include <linux/io.h>
+#include <linux/nd.h>
+#include "nd-private.h"
+#include "label.h"
+#include "nd.h"
+
+#include <asm-generic/io-64-nonatomic-lo-hi.h>
+
+static u32 best_seq(u32 a, u32 b)
+{
+ a &= NSINDEX_SEQ_MASK;
+ b &= NSINDEX_SEQ_MASK;
+
+ if (a == 0 || a == b)
+ return b;
+ else if (b == 0)
+ return a;
+ else if (nd_inc_seq(a) == b)
+ return b;
+ else
+ return a;
+}
+
+size_t sizeof_namespace_index(struct nd_dimm_drvdata *ndd)
+{
+ u32 index_span;
+
+ if (ndd->nsindex_size)
+ return ndd->nsindex_size;
+
+ /*
+ * The minimum index space is 512 bytes, with that amount of
+ * index we can describe ~1400 labels which is less than a byte
+ * of overhead per label. Round up to a byte of overhead per
+ * label and determine the size of the index region. Yes, this
+ * starts to waste space at larger config_sizes, but it's
+ * unlikely we'll ever see anything but 128K.
+ */
+ index_span = ndd->nsarea.config_size / 129;
+ index_span /= NSINDEX_ALIGN * 2;
+ ndd->nsindex_size = index_span * NSINDEX_ALIGN;
+
+ return ndd->nsindex_size;
+}
+
+int nd_label_validate(struct nd_dimm_drvdata *ndd)
+{
+ /*
+ * On media label format consists of two index blocks followed
+ * by an array of labels. None of these structures are ever
+ * updated in place. A sequence number tracks the current
+ * active index and the next one to write, while labels are
+ * written to free slots.
+ *
+ * +------------+
+ * | |
+ * | nsindex0 |
+ * | |
+ * +------------+
+ * | |
+ * | nsindex1 |
+ * | |
+ * +------------+
+ * | label0 |
+ * +------------+
+ * | label1 |
+ * +------------+
+ * | |
+ * ....nslot...
+ * | |
+ * +------------+
+ * | labelN |
+ * +------------+
+ */
+ struct nd_namespace_index __iomem *nsindex[] = {
+ to_namespace_index(ndd, 0),
+ to_namespace_index(ndd, 1),
+ };
+ const int num_index = ARRAY_SIZE(nsindex);
+ struct device *dev = ndd->dev;
+ bool valid[] = { false, false };
+ int i, num_valid = 0;
+ u32 seq;
+
+ for (i = 0; i < num_index; i++) {
+ u64 sum_save, sum;
+ u8 sig[NSINDEX_SIG_LEN];
+
+ memcpy_fromio(sig, nsindex[i]->sig, NSINDEX_SIG_LEN);
+ if (memcmp(sig, NSINDEX_SIGNATURE, NSINDEX_SIG_LEN) != 0) {
+ dev_dbg(dev, "%s: nsindex%d signature invalid\n",
+ __func__, i);
+ continue;
+ }
+ sum_save = readq(&nsindex[i]->checksum);
+ writeq(0, &nsindex[i]->checksum);
+ sum = nd_fletcher64(nsindex[i], sizeof_namespace_index(ndd));
+ writeq(sum_save, &nsindex[i]->checksum);
+ if (sum != sum_save) {
+ dev_dbg(dev, "%s: nsindex%d checksum invalid\n",
+ __func__, i);
+ continue;
+ }
+ if ((readl(&nsindex[i]->seq) & NSINDEX_SEQ_MASK) == 0) {
+ dev_dbg(dev, "%s: nsindex%d sequence: %#x invalid\n",
+ __func__, i, readl(&nsindex[i]->seq));
+ continue;
+ }
+
+ /* sanity check the index against expected values */
+ if (readq(&nsindex[i]->myoff)
+ != i * sizeof_namespace_index(ndd)) {
+ dev_dbg(dev, "%s: nsindex%d myoff: %#llx invalid\n",
+ __func__, i, (unsigned long long)
+ readq(&nsindex[i]->myoff));
+ continue;
+ }
+ if (readq(&nsindex[i]->otheroff)
+ != (!i) * sizeof_namespace_index(ndd)) {
+ dev_dbg(dev, "%s: nsindex%d otheroff: %#llx invalid\n",
+ __func__, i, (unsigned long long)
+ readq(&nsindex[i]->otheroff));
+ continue;
+ }
+ if (readq(&nsindex[i]->mysize) > sizeof_namespace_index(ndd)
+ || readq(&nsindex[i]->mysize)
+ < sizeof(struct nd_namespace_index)) {
+ dev_dbg(dev, "%s: nsindex%d mysize: %#llx invalid\n",
+ __func__, i, (unsigned long long)
+ readq(&nsindex[i]->mysize));
+ continue;
+ }
+ if (readl(&nsindex[i]->nslot) * sizeof(struct nd_namespace_label)
+ + 2 * sizeof_namespace_index(ndd)
+ > ndd->nsarea.config_size) {
+ dev_dbg(dev, "%s: nsindex%d nslot: %u invalid, config_size: %#x\n",
+ __func__, i, readl(&nsindex[i]->nslot),
+ ndd->nsarea.config_size);
+ continue;
+ }
+ valid[i] = true;
+ num_valid++;
+ }
+
+ switch (num_valid) {
+ case 0:
+ break;
+ case 1:
+ for (i = 0; i < num_index; i++)
+ if (valid[i])
+ return i;
+ /* can't have num_valid > 0 but valid[] = { false, false } */
+ WARN_ON(1);
+ break;
+ default:
+ /* pick the best index... */
+ seq = best_seq(readl(&nsindex[0]->seq), readl(&nsindex[1]->seq));
+ if (seq == (readl(&nsindex[1]->seq) & NSINDEX_SEQ_MASK))
+ return 1;
+ else
+ return 0;
+ break;
+ }
+
+ return -1;
+}
+
+void nd_label_copy(struct nd_dimm_drvdata *ndd,
+ struct nd_namespace_index __iomem *dst,
+ struct nd_namespace_index __iomem *src)
+{
+ void *s, *d;
+
+ if (dst && src)
+ /* pass */;
+ else
+ return;
+
+ d = (void * __force) dst;
+ s = (void * __force) src;
+ memcpy(d, s, sizeof_namespace_index(ndd));
+}
+
+static struct nd_namespace_label __iomem *nd_label_base(struct nd_dimm_drvdata *ndd)
+{
+ void *base = to_namespace_index(ndd, 0);
+
+ return base + 2 * sizeof_namespace_index(ndd);
+}
+
+#define for_each_clear_bit_le(bit, addr, size) \
+ for ((bit) = find_next_zero_bit_le((addr), (size), 0); \
+ (bit) < (size); \
+ (bit) = find_next_zero_bit_le((addr), (size), (bit) + 1))
+
+/**
+ * preamble_current - common variable initialization for nd_label_* routines
+ * @nd_dimm: dimm container for the relevant label set
+ * @nsindex: on return set to the currently active namespace index
+ * @free: on return set to the free label bitmap in the index
+ * @nslot: on return set to the number of slots in the label space
+ */
+static bool preamble_current(struct nd_dimm_drvdata *ndd,
+ struct nd_namespace_index **nsindex,
+ unsigned long **free, u32 *nslot)
+{
+ *nsindex = to_current_namespace_index(ndd);
+ if (*nsindex == NULL)
+ return false;
+
+ *free = (unsigned long __force *) (*nsindex)->free;
+ *nslot = readl(&(*nsindex)->nslot);
+
+ return true;
+}
+
+static char *nd_label_gen_id(struct nd_label_id *label_id, u8 *uuid, u32 flags)
+{
+ if (!label_id || !uuid)
+ return NULL;
+ snprintf(label_id->id, ND_LABEL_ID_SIZE, "%s-%pUb",
+ flags & NSLABEL_FLAG_LOCAL ? "blk" : "pmem", uuid);
+ return label_id->id;
+}
+
+static bool slot_valid(struct nd_namespace_label __iomem *nd_label, u32 slot)
+{
+ /* check that we are written where we expect to be written */
+ if (slot != readl(&nd_label->slot))
+ return false;
+
+ /* check that DPA allocations are page aligned */
+ if ((readq(&nd_label->dpa) | readq(&nd_label->rawsize)) % SZ_4K)
+ return false;
+
+ return true;
+}
+
+int nd_label_reserve_dpa(struct nd_dimm_drvdata *ndd)
+{
+ struct nd_namespace_index __iomem *nsindex;
+ unsigned long *free;
+ u32 nslot, slot;
+
+ if (!preamble_current(ndd, &nsindex, &free, &nslot))
+ return 0; /* no label, nothing to reserve */
+
+ for_each_clear_bit_le(slot, free, nslot) {
+ struct nd_namespace_label __iomem *nd_label;
+ struct nd_region *nd_region = NULL;
+ u8 label_uuid[NSLABEL_UUID_LEN];
+ struct nd_label_id *label_id;
+ struct resource *res;
+ u32 flags;
+
+ nd_label = nd_label_base(ndd) + slot;
+
+ if (!slot_valid(nd_label, slot))
+ continue;
+
+ label_id = devm_kzalloc(ndd->dev, sizeof(*label_id),
+ GFP_KERNEL);
+ if (!label_id)
+ return -ENOMEM;
+ memcpy_fromio(label_uuid, nd_label->uuid,
+ NSLABEL_UUID_LEN);
+ flags = readl(&nd_label->flags);
+ res = __request_region(&ndd->dpa, readq(&nd_label->dpa),
+ readq(&nd_label->rawsize),
+ nd_label_gen_id(label_id, label_uuid, flags), 0);
+ nd_dbg_dpa(nd_region, ndd, res, "reserve\n");
+ if (!res)
+ return -EBUSY;
+ }
+
+ return 0;
+}
diff --git a/drivers/block/nd/label.h b/drivers/block/nd/label.h
new file mode 100644
index 000000000000..79ed885a43c0
--- /dev/null
+++ b/drivers/block/nd/label.h
@@ -0,0 +1,129 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#ifndef __LABEL_H__
+#define __LABEL_H__
+
+#include <linux/ndctl.h>
+#include <linux/sizes.h>
+#include <linux/io.h>
+
+enum {
+ NSINDEX_SIG_LEN = 16,
+ NSINDEX_ALIGN = 256,
+ NSINDEX_SEQ_MASK = 0x3,
+ NSLABEL_UUID_LEN = 16,
+ NSLABEL_NAME_LEN = 64,
+ NSLABEL_FLAG_ROLABEL = 0x1, /* read-only label */
+ NSLABEL_FLAG_LOCAL = 0x2, /* DIMM-local namespace */
+ NSLABEL_FLAG_BTT = 0x4, /* namespace contains a BTT */
+ NSLABEL_FLAG_UPDATING = 0x8, /* label being updated */
+ BTT_ALIGN = 4096, /* all btt structures */
+ BTTINFO_SIG_LEN = 16,
+ BTTINFO_UUID_LEN = 16,
+ BTTINFO_FLAG_ERROR = 0x1, /* error state (read-only) */
+ BTTINFO_MAJOR_VERSION = 1,
+ ND_LABEL_MIN_SIZE = 512 * 129, /* see sizeof_namespace_index() */
+ ND_LABEL_ID_SIZE = 50,
+};
+
+static const char NSINDEX_SIGNATURE[] = "NAMESPACE_INDEX\0";
+
+/**
+ * struct nd_namespace_index - label set superblock
+ * @sig: NAMESPACE_INDEX\0
+ * @flags: placeholder
+ * @seq: sequence number for this index
+ * @myoff: offset of this index in label area
+ * @mysize: size of this index struct
+ * @otheroff: offset of other index
+ * @labeloff: offset of first label slot
+ * @nslot: total number of label slots
+ * @major: label area major version
+ * @minor: label area minor version
+ * @checksum: fletcher64 of all fields
+ * @free[0]: bitmap, nlabel bits
+ *
+ * The size of free[] is rounded up so the total struct size is a
+ * multiple of NSINDEX_ALIGN bytes. Any bits this allocates beyond
+ * nlabel bits must be zero.
+ */
+struct nd_namespace_index {
+ u8 sig[NSINDEX_SIG_LEN];
+ __le32 flags;
+ __le32 seq;
+ __le64 myoff;
+ __le64 mysize;
+ __le64 otheroff;
+ __le64 labeloff;
+ __le32 nslot;
+ __le16 major;
+ __le16 minor;
+ __le64 checksum;
+ u8 free[0];
+};
+
+/**
+ * struct nd_namespace_label - namespace superblock
+ * @uuid: UUID per RFC 4122
+ * @name: optional name (NULL-terminated)
+ * @flags: see NSLABEL_FLAG_*
+ * @nlabel: num labels to describe this ns
+ * @position: labels position in set
+ * @isetcookie: interleave set cookie
+ * @lbasize: LBA size in bytes or 0 for pmem
+ * @dpa: DPA of NVM range on this DIMM
+ * @rawsize: size of namespace
+ * @slot: slot of this label in label area
+ * @unused: must be zero
+ */
+struct nd_namespace_label {
+ u8 uuid[NSLABEL_UUID_LEN];
+ u8 name[NSLABEL_NAME_LEN];
+ __le32 flags;
+ __le16 nlabel;
+ __le16 position;
+ __le64 isetcookie;
+ __le64 lbasize;
+ __le64 dpa;
+ __le64 rawsize;
+ __le32 slot;
+ __le32 unused;
+};
+
+/**
+ * struct nd_label_id - identifier string for dpa allocation
+ * @id: "{blk|pmem}-<namespace uuid>"
+ */
+struct nd_label_id {
+ char id[ND_LABEL_ID_SIZE];
+};
+
+/*
+ * If the 'best' index is invalid, so is the 'next' index. Otherwise,
+ * the next index is MOD(index+1, 2)
+ */
+static inline int nd_label_next_nsindex(int index)
+{
+ if (index < 0)
+ return -1;
+
+ return (index + 1) % 2;
+}
+
+struct nd_dimm_drvdata;
+int nd_label_validate(struct nd_dimm_drvdata *ndd);
+void nd_label_copy(struct nd_dimm_drvdata *ndd,
+ struct nd_namespace_index *dst,
+ struct nd_namespace_index *src);
+size_t sizeof_namespace_index(struct nd_dimm_drvdata *ndd);
+#endif /* __LABEL_H__ */
diff --git a/drivers/block/nd/nd.h b/drivers/block/nd/nd.h
index c69707dbd272..832103a5e3f7 100644
--- a/drivers/block/nd/nd.h
+++ b/drivers/block/nd/nd.h
@@ -16,11 +16,15 @@
#include <linux/mutex.h>
#include <linux/ndctl.h>
#include "libnd.h"
+#include "label.h"
struct nd_dimm_drvdata {
struct device *dev;
+ int nsindex_size;
struct nd_cmd_get_config_size nsarea;
void *data;
+ int ns_current, ns_next;
+ struct resource dpa;
};
struct nd_region_namespaces {
@@ -28,6 +32,37 @@ struct nd_region_namespaces {
int active;
};
+static inline struct nd_namespace_index __iomem *to_namespace_index(
+ struct nd_dimm_drvdata *ndd, int i)
+{
+ if (i < 0)
+ return NULL;
+
+ return ((void __iomem *) ndd->data + sizeof_namespace_index(ndd) * i);
+}
+
+static inline struct nd_namespace_index __iomem *to_current_namespace_index(
+ struct nd_dimm_drvdata *ndd)
+{
+ return to_namespace_index(ndd, ndd->ns_current);
+}
+
+static inline struct nd_namespace_index __iomem *to_next_namespace_index(
+ struct nd_dimm_drvdata *ndd)
+{
+ return to_namespace_index(ndd, ndd->ns_next);
+}
+
+#define nd_dbg_dpa(r, d, res, fmt, arg...) \
+ dev_dbg((r) ? &(r)->dev : (d)->dev, "%s: %.13s: %#llx @ %#llx " fmt, \
+ (r) ? dev_name((d)->dev) : "", res ? res->name : "null", \
+ (unsigned long long) (res ? resource_size(res) : 0), \
+ (unsigned long long) (res ? res->start : 0), ##arg)
+
+#define for_each_dpa_resource_safe(ndd, res, next) \
+ for (res = (ndd)->dpa.child, next = res ? res->sibling : NULL; \
+ res; res = next, next = next ? next->sibling : NULL)
+
struct nd_region {
struct device dev;
u16 ndr_mappings;
@@ -39,6 +74,15 @@ struct nd_region {
struct nd_mapping mapping[0];
};
+/*
+ * Lookup next in the repeating sequence of 01, 10, and 11.
+ */
+static inline unsigned nd_inc_seq(unsigned seq)
+{
+ static const unsigned next[] = { 0, 2, 3, 1 };
+
+ return next[seq & 3];
+}
enum nd_async_mode {
ND_SYNC,
ND_ASYNC,
@@ -54,4 +98,5 @@ int nd_region_register_namespaces(struct nd_region *nd_region, int *err);
void nd_bus_lock(struct device *dev);
void nd_bus_unlock(struct device *dev);
bool is_nd_bus_locked(struct device *dev);
+int nd_label_reserve_dpa(struct nd_dimm_drvdata *ndd);
#endif /* __ND_H__ */
diff --git a/include/uapi/linux/ndctl.h b/include/uapi/linux/ndctl.h
index 5ffa319f3408..624a19d9e6e4 100644
--- a/include/uapi/linux/ndctl.h
+++ b/include/uapi/linux/ndctl.h
@@ -175,7 +175,6 @@ static inline const char *nd_dimm_cmd_name(unsigned cmd)
#define ND_IOCTL_ARS_QUERY _IOWR(ND_IOCTL, ND_CMD_ARS_QUERY,\
struct nd_cmd_ars_query)
-
#define ND_DEVICE_DIMM 1 /* nd_dimm: container for "config data" */
#define ND_DEVICE_REGION_PMEM 2 /* nd_region: (parent of pmem namespaces) */
#define ND_DEVICE_REGION_BLK 3 /* nd_region: (parent of blk namespaces) */
A complete label set is a PMEM-label per dimm where all the UUIDs
match and the interleave set cookie matches an active interleave set.
Present a sysfs ABI for manipulation of a PMEM-namespace's 'alt_name',
'uuid', and 'size' attributes. A later patch will make these settings
persistent by writing back the label.
Note that PMEM allocations grow forwards from the start of an interleave
set (lowest dimm-physical-address (DPA)). BLK-namespaces that alias
with a PMEM interleave set will grow allocations backward from the
highest DPA.
Cc: Greg KH <[email protected]>
Cc: Neil Brown <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/bus.c | 6
drivers/block/nd/core.c | 64 ++
drivers/block/nd/dimm.c | 2
drivers/block/nd/dimm_devs.c | 127 +++++
drivers/block/nd/label.c | 54 ++
drivers/block/nd/label.h | 3
drivers/block/nd/libnd.h | 2
drivers/block/nd/namespace_devs.c | 1024 +++++++++++++++++++++++++++++++++++++
drivers/block/nd/nd-private.h | 14 +
drivers/block/nd/nd.h | 33 +
drivers/block/nd/pmem.c | 22 +
drivers/block/nd/region_devs.c | 145 +++++
include/linux/nd.h | 24 +
include/uapi/linux/ndctl.h | 4
14 files changed, 1512 insertions(+), 12 deletions(-)
diff --git a/drivers/block/nd/bus.c b/drivers/block/nd/bus.c
index 8afb8d4a7e81..819259e92468 100644
--- a/drivers/block/nd/bus.c
+++ b/drivers/block/nd/bus.c
@@ -364,8 +364,10 @@ u32 nd_cmd_out_size(struct nd_dimm *nd_dimm, int cmd,
}
EXPORT_SYMBOL_GPL(nd_cmd_out_size);
-static void wait_nd_bus_probe_idle(struct nd_bus *nd_bus)
+void wait_nd_bus_probe_idle(struct device *dev)
{
+ struct nd_bus *nd_bus = walk_to_nd_bus(dev);
+
do {
if (nd_bus->probe_active == 0)
break;
@@ -384,7 +386,7 @@ static int nd_cmd_clear_to_send(struct nd_dimm *nd_dimm, unsigned int cmd)
return 0;
nd_bus = walk_to_nd_bus(&nd_dimm->dev);
- wait_nd_bus_probe_idle(nd_bus);
+ wait_nd_bus_probe_idle(&nd_bus->dev);
if (atomic_read(&nd_dimm->busy))
return -EBUSY;
diff --git a/drivers/block/nd/core.c b/drivers/block/nd/core.c
index 603970d0ef3a..cf64b7a50d3a 100644
--- a/drivers/block/nd/core.c
+++ b/drivers/block/nd/core.c
@@ -13,6 +13,7 @@
#include <linux/export.h>
#include <linux/module.h>
#include <linux/device.h>
+#include <linux/ctype.h>
#include <linux/ndctl.h>
#include <linux/mutex.h>
#include <linux/slab.h>
@@ -106,6 +107,69 @@ struct nd_bus *walk_to_nd_bus(struct device *nd_dev)
return NULL;
}
+static bool is_uuid_sep(char sep)
+{
+ if (sep == '\n' || sep == '-' || sep == ':' || sep == '\0')
+ return true;
+ return false;
+}
+
+static int nd_uuid_parse(struct device *dev, u8 *uuid_out, const char *buf,
+ size_t len)
+{
+ const char *str = buf;
+ u8 uuid[16];
+ int i;
+
+ for (i = 0; i < 16; i++) {
+ if (!isxdigit(str[0]) || !isxdigit(str[1])) {
+ dev_dbg(dev, "%s: pos: %d buf[%zd]: %c buf[%zd]: %c\n",
+ __func__, i, str - buf, str[0],
+ str + 1 - buf, str[1]);
+ return -EINVAL;
+ }
+
+ uuid[i] = (hex_to_bin(str[0]) << 4) | hex_to_bin(str[1]);
+ str += 2;
+ if (is_uuid_sep(*str))
+ str++;
+ }
+
+ memcpy(uuid_out, uuid, sizeof(uuid));
+ return 0;
+}
+
+/**
+ * nd_uuid_store: common implementation for writing 'uuid' sysfs attributes
+ * @dev: container device for the uuid property
+ * @uuid_out: uuid buffer to replace
+ * @buf: raw sysfs buffer to parse
+ *
+ * Enforce that uuids can only be changed while the device is disabled
+ * (driver detached)
+ * LOCKING: expects device_lock() is held on entry
+ */
+int nd_uuid_store(struct device *dev, u8 **uuid_out, const char *buf,
+ size_t len)
+{
+ u8 uuid[16];
+ int rc;
+
+ if (dev->driver)
+ return -EBUSY;
+
+ rc = nd_uuid_parse(dev, uuid, buf, len);
+ if (rc)
+ return rc;
+
+ kfree(*uuid_out);
+ *uuid_out = kmemdup(uuid, sizeof(uuid), GFP_KERNEL);
+ if (!(*uuid_out))
+ return -ENOMEM;
+
+ return 0;
+}
+
static ssize_t commands_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
diff --git a/drivers/block/nd/dimm.c b/drivers/block/nd/dimm.c
index 5477176c5de0..e2f964308672 100644
--- a/drivers/block/nd/dimm.c
+++ b/drivers/block/nd/dimm.c
@@ -86,7 +86,7 @@ static int nd_dimm_remove(struct device *dev)
nd_bus_lock(dev);
dev_set_drvdata(dev, NULL);
for_each_dpa_resource_safe(ndd, res, _r)
- __release_region(&ndd->dpa, res->start, resource_size(res));
+ nd_dimm_free_dpa(ndd, res);
nd_bus_unlock(dev);
free_data(ndd);
diff --git a/drivers/block/nd/dimm_devs.c b/drivers/block/nd/dimm_devs.c
index 3fbd0d0502eb..b242d3ae6d12 100644
--- a/drivers/block/nd/dimm_devs.c
+++ b/drivers/block/nd/dimm_devs.c
@@ -159,6 +159,14 @@ struct nd_dimm *to_nd_dimm(struct device *dev)
}
EXPORT_SYMBOL_GPL(to_nd_dimm);
+struct nd_dimm_drvdata *to_ndd(struct nd_mapping *nd_mapping)
+{
+ struct nd_dimm *nd_dimm = nd_mapping->nd_dimm;
+
+ return dev_get_drvdata(&nd_dimm->dev);
+}
+EXPORT_SYMBOL(to_ndd);
+
const char *nd_dimm_name(struct nd_dimm *nd_dimm)
{
return dev_name(&nd_dimm->dev);
@@ -247,6 +255,125 @@ struct nd_dimm *nd_dimm_create(struct nd_bus *nd_bus, void *provider_data,
}
EXPORT_SYMBOL_GPL(nd_dimm_create);
+/**
+ * nd_pmem_available_dpa - for the given dimm+region account unallocated dpa
+ * @nd_mapping: container of dpa-resource-root + labels
+ * @nd_region: constrain available space check to this reference region
+ * @overlap: calculate available space assuming this level of overlap
+ *
+ * Validate that a PMEM label, if present, aligns with the start of an
+ * interleave set and truncate the available size at the lowest BLK
+ * overlap point.
+ *
+ * The expectation is that this routine is called multiple times as it
+ * probes for the largest BLK encroachment for any single member DIMM of
+ * the interleave set. Once that value is determined the PMEM-limit for
+ * the set can be established.
+ */
+resource_size_t nd_pmem_available_dpa(struct nd_region *nd_region,
+ struct nd_mapping *nd_mapping, resource_size_t *overlap)
+{
+ resource_size_t map_end, busy = 0, available, blk_start;
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ struct resource *res;
+ const char *reason;
+
+ if (!ndd)
+ return 0;
+
+ map_end = nd_mapping->start + nd_mapping->size - 1;
+ blk_start = max(nd_mapping->start, map_end + 1 - *overlap);
+ for_each_dpa_resource(ndd, res)
+ if (res->start >= nd_mapping->start && res->start < map_end) {
+ if (strncmp(res->name, "blk", 3) == 0)
+ blk_start = min(blk_start, res->start);
+ else if (res->start != nd_mapping->start) {
+ reason = "misaligned to iset";
+ goto err;
+ } else {
+ if (busy) {
+ reason = "duplicate overlapping PMEM reservations?";
+ goto err;
+ }
+ busy += resource_size(res);
+ continue;
+ }
+ } else if (res->end >= nd_mapping->start && res->end <= map_end) {
+ if (strncmp(res->name, "blk", 3) == 0) {
+ /*
+ * If a BLK allocation overlaps the start of
+ * PMEM the entire interleave set may now only
+ * be used for BLK.
+ */
+ blk_start = nd_mapping->start;
+ } else {
+ reason = "misaligned to iset";
+ goto err;
+ }
+ } else if (nd_mapping->start > res->start
+ && nd_mapping->start < res->end) {
+ /* total eclipse of the mapping */
+ busy += nd_mapping->size;
+ blk_start = nd_mapping->start;
+ }
+
+ *overlap = map_end + 1 - blk_start;
+ available = blk_start - nd_mapping->start;
+ if (busy < available)
+ return available - busy;
+ return 0;
+
+ err:
+ /*
+ * Something is wrong, PMEM must align with the start of the
+ * interleave set, and there can only be one allocation per set.
+ */
+ nd_dbg_dpa(nd_region, ndd, res, "%s\n", reason);
+ return 0;
+}
+
+void nd_dimm_free_dpa(struct nd_dimm_drvdata *ndd, struct resource *res)
+{
+ WARN_ON_ONCE(!is_nd_bus_locked(ndd->dev));
+ kfree(res->name);
+ __release_region(&ndd->dpa, res->start, resource_size(res));
+}
+
+struct resource *nd_dimm_allocate_dpa(struct nd_dimm_drvdata *ndd,
+ struct nd_label_id *label_id, resource_size_t start,
+ resource_size_t n)
+{
+ char *name = kmemdup(label_id, sizeof(*label_id), GFP_KERNEL);
+ struct resource *res;
+
+ if (!name)
+ return NULL;
+
+ WARN_ON_ONCE(!is_nd_bus_locked(ndd->dev));
+ res = __request_region(&ndd->dpa, start, n, name, 0);
+ if (!res)
+ kfree(name);
+ return res;
+}
+
+/**
+ * nd_dimm_allocated_dpa - sum up the dpa currently allocated to this label_id
+ * @nd_dimm: container of dpa-resource-root + labels
+ * @label_id: dpa resource name of the form {pmem|blk}-<human readable uuid>
+ */
+resource_size_t nd_dimm_allocated_dpa(struct nd_dimm_drvdata *ndd,
+ struct nd_label_id *label_id)
+{
+ resource_size_t allocated = 0;
+ struct resource *res;
+
+ for_each_dpa_resource(ndd, res)
+ if (strcmp(res->name, label_id->id) == 0)
+ allocated += resource_size(res);
+
+ return allocated;
+}
+
static int count_dimms(struct device *dev, void *c)
{
int *count = c;
diff --git a/drivers/block/nd/label.c b/drivers/block/nd/label.c
index e791ea8bbdde..b55fa2a6f872 100644
--- a/drivers/block/nd/label.c
+++ b/drivers/block/nd/label.c
@@ -228,7 +228,7 @@ static bool preamble_current(struct nd_dimm_drvdata *ndd,
return true;
}
-static char *nd_label_gen_id(struct nd_label_id *label_id, u8 *uuid, u32 flags)
+char *nd_label_gen_id(struct nd_label_id *label_id, u8 *uuid, u32 flags)
{
if (!label_id || !uuid)
return NULL;
@@ -289,3 +289,55 @@ int nd_label_reserve_dpa(struct nd_dimm_drvdata *ndd)
return 0;
}
+
+int nd_label_active_count(struct nd_dimm_drvdata *ndd)
+{
+ struct nd_namespace_index __iomem *nsindex;
+ unsigned long *free;
+ u32 nslot, slot;
+ int count = 0;
+
+ if (!preamble_current(ndd, &nsindex, &free, &nslot))
+ return 0;
+
+ for_each_clear_bit_le(slot, free, nslot) {
+ struct nd_namespace_label __iomem *nd_label;
+
+ nd_label = nd_label_base(ndd) + slot;
+
+ if (!slot_valid(nd_label, slot)) {
+ dev_dbg(ndd->dev,
+ "%s: slot%d invalid slot: %d dpa: %lx rawsize: %lx\n",
+ __func__, slot, readl(&nd_label->slot),
+ (unsigned long) readq(&nd_label->dpa),
+ (unsigned long) readq(&nd_label->rawsize));
+ continue;
+ }
+ count++;
+ }
+ return count;
+}
+
+struct nd_namespace_label __iomem *nd_label_active(
+ struct nd_dimm_drvdata *ndd, int n)
+{
+ struct nd_namespace_index __iomem *nsindex;
+ unsigned long *free;
+ u32 nslot, slot;
+
+ if (!preamble_current(ndd, &nsindex, &free, &nslot))
+ return NULL;
+
+ for_each_clear_bit_le(slot, free, nslot) {
+ struct nd_namespace_label __iomem *nd_label;
+
+ nd_label = nd_label_base(ndd) + slot;
+ if (slot != readl(&nd_label->slot))
+ continue;
+
+ if (n-- == 0)
+ return nd_label_base(ndd) + slot;
+ }
+
+ return NULL;
+}
diff --git a/drivers/block/nd/label.h b/drivers/block/nd/label.h
index 79ed885a43c0..4436624f4146 100644
--- a/drivers/block/nd/label.h
+++ b/drivers/block/nd/label.h
@@ -126,4 +126,7 @@ void nd_label_copy(struct nd_dimm_drvdata *ndd,
struct nd_namespace_index *dst,
struct nd_namespace_index *src);
size_t sizeof_namespace_index(struct nd_dimm_drvdata *ndd);
+int nd_label_active_count(struct nd_dimm_drvdata *ndd);
+struct nd_namespace_label __iomem *nd_label_active(
+ struct nd_dimm_drvdata *ndd, int n);
#endif /* __LABEL_H__ */
diff --git a/drivers/block/nd/libnd.h b/drivers/block/nd/libnd.h
index deb29eff5b61..832dcfebbb49 100644
--- a/drivers/block/nd/libnd.h
+++ b/drivers/block/nd/libnd.h
@@ -40,8 +40,10 @@ typedef int (*ndctl_fn)(struct nd_bus_descriptor *nd_desc,
struct nd_dimm *nd_dimm, unsigned int cmd, void *buf,
unsigned int buf_len);
+struct nd_namespace_label;
struct nd_mapping {
struct nd_dimm *nd_dimm;
+ struct nd_namespace_label **labels;
u64 start;
u64 size;
};
diff --git a/drivers/block/nd/namespace_devs.c b/drivers/block/nd/namespace_devs.c
index 6861327f4245..fb23904df4c7 100644
--- a/drivers/block/nd/namespace_devs.c
+++ b/drivers/block/nd/namespace_devs.c
@@ -14,8 +14,11 @@
#include <linux/device.h>
#include <linux/slab.h>
#include <linux/nd.h>
+#include "nd-private.h"
#include "nd.h"
+#include <asm-generic/io-64-nonatomic-lo-hi.h>
+
static void namespace_io_release(struct device *dev)
{
struct nd_namespace_io *nsio = to_nd_namespace_io(dev);
@@ -23,11 +26,50 @@ static void namespace_io_release(struct device *dev)
kfree(nsio);
}
+static void namespace_pmem_release(struct device *dev)
+{
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+
+ kfree(nspm->alt_name);
+ kfree(nspm->uuid);
+ kfree(nspm);
+}
+
+static void namespace_blk_release(struct device *dev)
+{
+ /* TODO: blk namespace support */
+}
+
static struct device_type namespace_io_device_type = {
.name = "nd_namespace_io",
.release = namespace_io_release,
};
+static struct device_type namespace_pmem_device_type = {
+ .name = "nd_namespace_pmem",
+ .release = namespace_pmem_release,
+};
+
+static struct device_type namespace_blk_device_type = {
+ .name = "nd_namespace_blk",
+ .release = namespace_blk_release,
+};
+
+static bool is_namespace_pmem(struct device *dev)
+{
+ return dev ? dev->type == &namespace_pmem_device_type : false;
+}
+
+static bool is_namespace_blk(struct device *dev)
+{
+ return dev ? dev->type == &namespace_blk_device_type : false;
+}
+
+static bool is_namespace_io(struct device *dev)
+{
+ return dev ? dev->type == &namespace_io_device_type : false;
+}
+
static ssize_t type_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
@@ -37,13 +79,674 @@ static ssize_t type_show(struct device *dev,
}
static DEVICE_ATTR_RO(type);
+static ssize_t __alt_name_store(struct device *dev, const char *buf,
+ const size_t len)
+{
+ char *input, *pos, *alt_name, **ns_altname;
+ ssize_t rc;
+
+ if (is_namespace_pmem(dev)) {
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+
+ ns_altname = &nspm->alt_name;
+ } else if (is_namespace_blk(dev)) {
+ /* TODO: blk namespace support */
+ return -ENXIO;
+ } else
+ return -ENXIO;
+
+ if (dev->driver)
+ return -EBUSY;
+
+ input = kmemdup(buf, len + 1, GFP_KERNEL);
+ if (!input)
+ return -ENOMEM;
+
+ input[len] = '\0';
+ pos = strim(input);
+ if (strlen(pos) + 1 > NSLABEL_NAME_LEN) {
+ rc = -EINVAL;
+ goto out;
+ }
+
+ alt_name = kzalloc(NSLABEL_NAME_LEN, GFP_KERNEL);
+ if (!alt_name) {
+ rc = -ENOMEM;
+ goto out;
+ }
+ kfree(*ns_altname);
+ *ns_altname = alt_name;
+ sprintf(*ns_altname, "%s", pos);
+ rc = len;
+
+out:
+ kfree(input);
+ return rc;
+}
+
+static ssize_t alt_name_store(struct device *dev,
+ struct device_attribute *attr, const char *buf, size_t len)
+{
+ ssize_t rc;
+
+ device_lock(dev);
+ nd_bus_lock(dev);
+ wait_nd_bus_probe_idle(dev);
+ rc = __alt_name_store(dev, buf, len);
+ dev_dbg(dev, "%s: %s (%zd)\n", __func__, rc < 0 ? "fail" : "success", rc);
+ nd_bus_unlock(dev);
+ device_unlock(dev);
+
+ return rc;
+}
+
+static ssize_t alt_name_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ char *ns_altname;
+
+ if (is_namespace_pmem(dev)) {
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+
+ ns_altname = nspm->alt_name;
+ } else if (is_namespace_blk(dev)) {
+ /* TODO: blk namespace support */
+ return -ENXIO;
+ } else
+ return -ENXIO;
+
+ return sprintf(buf, "%s\n", ns_altname ? ns_altname : "");
+}
+static DEVICE_ATTR_RW(alt_name);
+
+static int scan_free(struct nd_region *nd_region,
+ struct nd_mapping *nd_mapping, struct nd_label_id *label_id,
+ resource_size_t n)
+{
+ bool is_blk = strncmp(label_id->id, "blk", 3) == 0;
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ int rc = 0;
+
+ while (n) {
+ struct resource *res, *last;
+ resource_size_t new_start;
+
+ last = NULL;
+ for_each_dpa_resource(ndd, res)
+ if (strcmp(res->name, label_id->id) == 0)
+ last = res;
+ res = last;
+ if (!res)
+ return 0;
+
+ if (n >= resource_size(res)) {
+ n -= resource_size(res);
+ nd_dbg_dpa(nd_region, ndd, res, "delete %d\n", rc);
+ nd_dimm_free_dpa(ndd, res);
+ /* retry with last resource deleted */
+ continue;
+ }
+
+ /*
+ * Keep BLK allocations relegated to high DPA as much as
+ * possible
+ */
+ if (is_blk)
+ new_start = res->start + n;
+ else
+ new_start = res->start;
+
+ rc = adjust_resource(res, new_start, resource_size(res) - n);
+ nd_dbg_dpa(nd_region, ndd, res, "shrink %d\n", rc);
+ break;
+ }
+
+ return rc;
+}
+
+/**
+ * shrink_dpa_allocation - for each dimm in region free n bytes for label_id
+ * @nd_region: the set of dimms to reclaim @n bytes from
+ * @label_id: unique identifier for the namespace consuming this dpa range
+ * @n: number of bytes per-dimm to release
+ *
+ * Assumes resources are ordered. Starting from the end try to
+ * adjust_resource() the allocation to @n, but if @n is larger than the
+ * allocation delete it and find the 'new' last allocation in the label
+ * set.
+ */
+static int shrink_dpa_allocation(struct nd_region *nd_region,
+ struct nd_label_id *label_id, resource_size_t n)
+{
+ int i;
+
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ int rc;
+
+ rc = scan_free(nd_region, nd_mapping, label_id, n);
+ if (rc)
+ return rc;
+ }
+
+ return 0;
+}
+
+static resource_size_t init_dpa_allocation(struct nd_label_id *label_id,
+ struct nd_region *nd_region, struct nd_mapping *nd_mapping,
+ resource_size_t n)
+{
+ bool is_blk = strncmp(label_id->id, "blk", 3) == 0;
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ resource_size_t first_dpa;
+ struct resource *res;
+ int rc = 0;
+
+ /* allocate blk from highest dpa first */
+ if (is_blk)
+ first_dpa = nd_mapping->start + nd_mapping->size - n;
+ else
+ first_dpa = nd_mapping->start;
+
+ /* first resource allocation for this label-id or dimm */
+ res = nd_dimm_allocate_dpa(ndd, label_id, first_dpa, n);
+ if (!res)
+ rc = -EBUSY;
+
+ nd_dbg_dpa(nd_region, ndd, res, "init %d\n", rc);
+ return rc ? n : 0;
+}
+
+static bool space_valid(bool is_pmem, struct nd_label_id *label_id,
+ struct resource *res)
+{
+ /*
+ * For BLK-space any space is valid, for PMEM-space, it must be
+ * contiguous with an existing allocation.
+ */
+ if (!is_pmem)
+ return true;
+ if (!res || strcmp(res->name, label_id->id) == 0)
+ return true;
+ return false;
+}
+
+enum alloc_loc {
+ ALLOC_ERR = 0, ALLOC_BEFORE, ALLOC_MID, ALLOC_AFTER,
+};
+
+static resource_size_t scan_allocate(struct nd_region *nd_region,
+ struct nd_mapping *nd_mapping, struct nd_label_id *label_id,
+ resource_size_t n)
+{
+ resource_size_t mapping_end = nd_mapping->start + nd_mapping->size - 1;
+ bool is_pmem = strncmp(label_id->id, "pmem", 4) == 0;
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ const resource_size_t to_allocate = n;
+ struct resource *res;
+ int first;
+
+ retry:
+ first = 0;
+ for_each_dpa_resource(ndd, res) {
+ resource_size_t allocate, available = 0, free_start, free_end;
+ struct resource *next = res->sibling, *new_res = NULL;
+ enum alloc_loc loc = ALLOC_ERR;
+ const char *action;
+ int rc = 0;
+
+ /* ignore resources outside this nd_mapping */
+ if (res->start > mapping_end)
+ continue;
+ if (res->end < nd_mapping->start)
+ continue;
+
+ /* space at the beginning of the mapping */
+ if (!first++ && res->start > nd_mapping->start) {
+ free_start = nd_mapping->start;
+ available = res->start - free_start;
+ if (space_valid(is_pmem, label_id, NULL))
+ loc = ALLOC_BEFORE;
+ }
+
+ /* space between allocations */
+ if (!loc && next) {
+ free_start = res->start + resource_size(res);
+ free_end = min(mapping_end, next->start - 1);
+ if (space_valid(is_pmem, label_id, res)
+ && free_start < free_end) {
+ available = free_end + 1 - free_start;
+ loc = ALLOC_MID;
+ }
+ }
+
+ /* space at the end of the mapping */
+ if (!loc && !next) {
+ free_start = res->start + resource_size(res);
+ free_end = mapping_end;
+ if (space_valid(is_pmem, label_id, res)
+ && free_start < free_end) {
+ available = free_end + 1 - free_start;
+ loc = ALLOC_AFTER;
+ }
+ }
+
+ if (!loc || !available)
+ continue;
+ allocate = min(available, n);
+ switch (loc) {
+ case ALLOC_BEFORE:
+ if (strcmp(res->name, label_id->id) == 0) {
+ /* adjust current resource up */
+ if (is_pmem)
+ return n;
+ rc = adjust_resource(res, res->start - allocate,
+ resource_size(res) + allocate);
+ action = "cur grow up";
+ } else
+ action = "allocate";
+ break;
+ case ALLOC_MID:
+ if (strcmp(next->name, label_id->id) == 0) {
+ /* adjust next resource up */
+ if (is_pmem)
+ return n;
+ rc = adjust_resource(next, next->start
+ - allocate, resource_size(next)
+ + allocate);
+ new_res = next;
+ action = "next grow up";
+ } else if (strcmp(res->name, label_id->id) == 0) {
+ action = "grow down";
+ } else
+ action = "allocate";
+ break;
+ case ALLOC_AFTER:
+ if (strcmp(res->name, label_id->id) == 0)
+ action = "grow down";
+ else
+ action = "allocate";
+ break;
+ default:
+ return n;
+ }
+
+ if (strcmp(action, "allocate") == 0) {
+ /* BLK allocate bottom up */
+ if (!is_pmem)
+ free_start += available - allocate;
+ else if (free_start != nd_mapping->start)
+ return n;
+
+ new_res = nd_dimm_allocate_dpa(ndd, label_id,
+ free_start, allocate);
+ if (!new_res)
+ rc = -EBUSY;
+ } else if (strcmp(action, "grow down") == 0) {
+ /* adjust current resource down */
+ rc = adjust_resource(res, res->start, resource_size(res)
+ + allocate);
+ }
+
+ if (!new_res)
+ new_res = res;
+
+ nd_dbg_dpa(nd_region, ndd, new_res, "%s(%d) %d\n",
+ action, loc, rc);
+
+ if (rc)
+ return n;
+
+ n -= allocate;
+ if (n) {
+ /*
+ * Retry scan with newly inserted resources.
+ * For example, if we did an ALLOC_BEFORE
+ * insertion there may also have been space
+ * available for an ALLOC_AFTER insertion, so we
+ * need to check this same resource again
+ */
+ goto retry;
+ } else
+ return 0;
+ }
+
+ if (is_pmem && n == to_allocate)
+ return init_dpa_allocation(label_id, nd_region, nd_mapping, n);
+ return n;
+}
+
+/**
+ * grow_dpa_allocation - for each dimm allocate n bytes for @label_id
+ * @nd_region: the set of dimms to allocate @n more bytes from
+ * @label_id: unique identifier for the namespace consuming this dpa range
+ * @n: number of bytes per-dimm to add to the existing allocation
+ *
+ * Assumes resources are ordered. For BLK regions, first consume
+ * BLK-only available DPA free space, then consume PMEM-aliased DPA
+ * space starting at the highest DPA. For PMEM regions start
+ * allocations from the start of an interleave set and end at the first
+ * BLK allocation or the end of the interleave set, whichever comes
+ * first.
+ */
+static int grow_dpa_allocation(struct nd_region *nd_region,
+ struct nd_label_id *label_id, resource_size_t n)
+{
+ int i;
+
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ int rc;
+
+ rc = scan_allocate(nd_region, nd_mapping, label_id, n);
+ if (rc)
+ return rc;
+ }
+
+ return 0;
+}
+
+static void nd_namespace_pmem_set_size(struct nd_region *nd_region,
+ struct nd_namespace_pmem *nspm, resource_size_t size)
+{
+ struct resource *res = &nspm->nsio.res;
+
+ res->start = nd_region->ndr_start;
+ res->end = nd_region->ndr_start + size - 1;
+}
+
+static ssize_t __size_store(struct device *dev, unsigned long long val)
+{
+ resource_size_t allocated = 0, available = 0;
+ struct nd_region *nd_region = to_nd_region(dev->parent);
+ struct nd_mapping *nd_mapping;
+ struct nd_dimm_drvdata *ndd;
+ struct nd_label_id label_id;
+ u32 flags = 0, remainder;
+ u8 *uuid = NULL;
+ int rc, i;
+
+ if (dev->driver)
+ return -EBUSY;
+
+ if (is_namespace_pmem(dev)) {
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+
+ uuid = nspm->uuid;
+ } else if (is_namespace_blk(dev)) {
+ /* TODO: blk namespace support */
+ return -ENXIO;
+ }
+
+ /*
+ * We need a uuid for the allocation-label and dimm(s) on which
+ * to store the label.
+ */
+ if (!uuid || nd_region->ndr_mappings == 0)
+ return -ENXIO;
+
+ div_u64_rem(val, SZ_4K * nd_region->ndr_mappings, &remainder);
+ if (remainder) {
+ dev_dbg(dev, "%llu is not %dK aligned\n", val,
+ (SZ_4K * nd_region->ndr_mappings) / SZ_1K);
+ return -EINVAL;
+ }
+
+ nd_label_gen_id(&label_id, uuid, flags);
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ nd_mapping = &nd_region->mapping[i];
+ ndd = to_ndd(nd_mapping);
+
+ /*
+ * All dimms in an interleave set, or the base dimm for a blk
+ * region, need to be enabled for the size to be changed.
+ */
+ if (!ndd)
+ return -ENXIO;
+
+ allocated += nd_dimm_allocated_dpa(ndd, &label_id);
+ }
+ available = nd_region_available_dpa(nd_region);
+
+ if (val > available + allocated)
+ return -ENOSPC;
+
+ if (val == allocated)
+ return 0;
+
+ val = div_u64(val, nd_region->ndr_mappings);
+ allocated = div_u64(allocated, nd_region->ndr_mappings);
+ if (val < allocated)
+ rc = shrink_dpa_allocation(nd_region, &label_id, allocated - val);
+ else
+ rc = grow_dpa_allocation(nd_region, &label_id, val - allocated);
+
+ if (rc)
+ return rc;
+
+ if (is_namespace_pmem(dev)) {
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+
+ nd_namespace_pmem_set_size(nd_region, nspm,
+ val * nd_region->ndr_mappings);
+ }
+
+ return rc;
+}
+
+static ssize_t size_store(struct device *dev,
+ struct device_attribute *attr, const char *buf, size_t len)
+{
+ unsigned long long val;
+ u8 **uuid = NULL;
+ int rc;
+
+ rc = kstrtoull(buf, 0, &val);
+ if (rc)
+ return rc;
+
+ device_lock(dev);
+ nd_bus_lock(dev);
+ wait_nd_bus_probe_idle(dev);
+ rc = __size_store(dev, val);
+
+ if (is_namespace_pmem(dev)) {
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+
+ uuid = &nspm->uuid;
+ } else if (is_namespace_blk(dev)) {
+ /* TODO: blk namespace support */
+ rc = -ENXIO;
+ }
+
+ if (rc == 0 && val == 0 && uuid) {
+ /* setting size zero == 'delete namespace' */
+ kfree(*uuid);
+ *uuid = NULL;
+ }
+
+ dev_dbg(dev, "%s: %llx %s (%d)\n", __func__, val, rc < 0
+ ? "fail" : "success", rc);
+
+ nd_bus_unlock(dev);
+ device_unlock(dev);
+
+ return rc ? rc : len;
+}
+
+static ssize_t size_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ if (is_namespace_pmem(dev)) {
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+
+ return sprintf(buf, "%llu\n", (unsigned long long)
+ resource_size(&nspm->nsio.res));
+ } else if (is_namespace_blk(dev)) {
+ /* TODO: blk namespace support */
+ return -ENXIO;
+ } else if (is_namespace_io(dev)) {
+ struct nd_namespace_io *nsio = to_nd_namespace_io(dev);
+
+ return sprintf(buf, "%llu\n", (unsigned long long)
+ resource_size(&nsio->res));
+ } else
+ return -ENXIO;
+}
+static DEVICE_ATTR(size, S_IRUGO, size_show, size_store);
+
+static ssize_t uuid_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ u8 *uuid;
+
+ if (is_namespace_pmem(dev)) {
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+
+ uuid = nspm->uuid;
+ } else if (is_namespace_blk(dev)) {
+ /* TODO: blk namespace support */
+ return -ENXIO;
+ } else
+ return -ENXIO;
+
+ if (uuid)
+ return sprintf(buf, "%pUb\n", uuid);
+ return sprintf(buf, "\n");
+}
+
+/**
+ * namespace_update_uuid - check for a unique uuid and whether we're "renaming"
+ * @nd_region: parent region so we can updates all dimms in the set
+ * @dev: namespace type for generating label_id
+ * @new_uuid: incoming uuid
+ * @old_uuid: reference to the uuid storage location in the namespace object
+ */
+static int namespace_update_uuid(struct nd_region *nd_region,
+ struct device *dev, u8 *new_uuid, u8 **old_uuid)
+{
+ u32 flags = is_namespace_blk(dev) ? NSLABEL_FLAG_LOCAL : 0;
+ struct nd_label_id old_label_id;
+ struct nd_label_id new_label_id;
+ int i, rc;
+
+ rc = nd_is_uuid_unique(dev, new_uuid) ? 0 : -EINVAL;
+ if (rc) {
+ kfree(new_uuid);
+ return rc;
+ }
+
+ if (*old_uuid == NULL)
+ goto out;
+
+ nd_label_gen_id(&old_label_id, *old_uuid, flags);
+ nd_label_gen_id(&new_label_id, new_uuid, flags);
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ struct resource *res;
+
+ for_each_dpa_resource(ndd, res)
+ if (strcmp(res->name, old_label_id.id) == 0)
+ sprintf((void *) res->name, "%s",
+ new_label_id.id);
+ }
+ kfree(*old_uuid);
+ out:
+ *old_uuid = new_uuid;
+ return 0;
+}
+
+static ssize_t uuid_store(struct device *dev,
+ struct device_attribute *attr, const char *buf, size_t len)
+{
+ struct nd_region *nd_region = to_nd_region(dev->parent);
+ u8 *uuid = NULL;
+ u8 **ns_uuid;
+ ssize_t rc;
+
+ if (is_namespace_pmem(dev)) {
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+
+ ns_uuid = &nspm->uuid;
+ } else if (is_namespace_blk(dev)) {
+ /* TODO: blk namespace support */
+ return -ENXIO;
+ } else
+ return -ENXIO;
+
+ device_lock(dev);
+ nd_bus_lock(dev);
+ wait_nd_bus_probe_idle(dev);
+ rc = nd_uuid_store(dev, &uuid, buf, len);
+ if (rc >= 0)
+ rc = namespace_update_uuid(nd_region, dev, uuid, ns_uuid);
+ dev_dbg(dev, "%s: result: %zd wrote: %s%s", __func__,
+ rc, buf, buf[len - 1] == '\n' ? "" : "\n");
+ nd_bus_unlock(dev);
+ device_unlock(dev);
+
+ return rc ? rc : len;
+}
+static DEVICE_ATTR_RW(uuid);
+
+static ssize_t resource_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct resource *res;
+
+ if (is_namespace_pmem(dev)) {
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+
+ res = &nspm->nsio.res;
+ } else if (is_namespace_io(dev)) {
+ struct nd_namespace_io *nsio = to_nd_namespace_io(dev);
+
+ res = &nsio->res;
+ } else
+ return -ENXIO;
+
+ /* no address to convey if the namespace has no allocation */
+ if (resource_size(res) == 0)
+ return -ENXIO;
+ return sprintf(buf, "%#llx\n", (unsigned long long) res->start);
+}
+static DEVICE_ATTR_RO(resource);
+
static struct attribute *nd_namespace_attributes[] = {
&dev_attr_type.attr,
+ &dev_attr_size.attr,
+ &dev_attr_uuid.attr,
+ &dev_attr_resource.attr,
+ &dev_attr_alt_name.attr,
NULL,
};
+static umode_t nd_namespace_attr_visible(struct kobject *kobj, struct attribute *a, int n)
+{
+ struct device *dev = container_of(kobj, struct device, kobj);
+
+ if (a == &dev_attr_resource.attr) {
+ if (is_namespace_blk(dev))
+ return 0;
+ return a->mode;
+ }
+
+ if (is_namespace_pmem(dev) || is_namespace_blk(dev)) {
+ if (a == &dev_attr_size.attr)
+ return S_IWUSR;
+ return a->mode;
+ }
+
+ if (a == &dev_attr_type.attr || a == &dev_attr_size.attr)
+ return a->mode;
+
+ return 0;
+}
+
static struct attribute_group nd_namespace_attribute_group = {
.attrs = nd_namespace_attributes,
+ .is_visible = nd_namespace_attr_visible,
};
static const struct attribute_group *nd_namespace_attribute_groups[] = {
@@ -80,23 +783,326 @@ static struct device **create_namespace_io(struct nd_region *nd_region)
return devs;
}
+static bool has_uuid_at_pos(struct nd_region *nd_region, u8 *uuid, u64 cookie, u16 pos)
+{
+ struct nd_namespace_label __iomem *found = NULL;
+ int i;
+
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ struct nd_namespace_label __iomem *nd_label;
+ u8 label_uuid[NSLABEL_UUID_LEN];
+ u8 *found_uuid = NULL;
+ int l;
+
+ for_each_label(l, nd_label, nd_mapping->labels) {
+ u64 isetcookie = readq(&nd_label->isetcookie);
+ u16 position = readw(&nd_label->position);
+ u16 nlabel = readw(&nd_label->nlabel);
+
+ if (isetcookie != cookie)
+ continue;
+
+ memcpy_fromio(label_uuid, nd_label->uuid,
+ NSLABEL_UUID_LEN);
+ if (memcmp(label_uuid, uuid, NSLABEL_UUID_LEN) != 0)
+ continue;
+
+ if (found_uuid) {
+ dev_dbg(to_ndd(nd_mapping)->dev,
+ "%s duplicate entry for uuid\n",
+ __func__);
+ return false;
+ }
+ found_uuid = label_uuid;
+ if (nlabel != nd_region->ndr_mappings)
+ continue;
+ if (position != pos)
+ continue;
+ found = nd_label;
+ break;
+ }
+ if (found)
+ break;
+ }
+ return found != NULL;
+}
+
+static int select_pmem_uuid(struct nd_region *nd_region, u8 *pmem_uuid)
+{
+ struct nd_namespace_label __iomem *select = NULL;
+ int i;
+
+ if (!pmem_uuid)
+ return -ENODEV;
+
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ struct nd_namespace_label __iomem *nd_label;
+ u64 hw_start, hw_end, pmem_start, pmem_end;
+ int l;
+
+ for_each_label(l, nd_label, nd_mapping->labels) {
+ u8 label_uuid[NSLABEL_UUID_LEN];
+
+ memcpy_fromio(label_uuid, nd_label->uuid,
+ NSLABEL_UUID_LEN);
+ if (memcmp(label_uuid, pmem_uuid, NSLABEL_UUID_LEN) == 0)
+ break;
+ }
+
+ if (!nd_label) {
+ WARN_ON(1);
+ return -EINVAL;
+ }
+
+ select = nd_label;
+ /*
+ * Check that this label is compliant with the dpa
+ * range published in NFIT
+ */
+ hw_start = nd_mapping->start;
+ hw_end = hw_start + nd_mapping->size;
+ pmem_start = readq(&select->dpa);
+ pmem_end = pmem_start + readq(&select->rawsize);
+ if (pmem_start == hw_start && pmem_end <= hw_end)
+ /* pass */;
+ else
+ return -EINVAL;
+
+ nd_set_label(nd_mapping->labels, select, 0);
+ nd_set_label(nd_mapping->labels, (void __iomem *) NULL, 1);
+ }
+ return 0;
+}
+
+/**
+ * find_pmem_label_set - validate interleave set labelling, retrieve label0
+ * @nd_region: region with mappings to validate
+ */
+static int find_pmem_label_set(struct nd_region *nd_region,
+ struct nd_namespace_pmem *nspm)
+{
+ u64 cookie = nd_region_interleave_set_cookie(nd_region);
+ struct nd_namespace_label __iomem *nd_label;
+ u8 select_uuid[NSLABEL_UUID_LEN];
+ resource_size_t size = 0;
+ u8 *pmem_uuid = NULL;
+ int rc = -ENODEV, l;
+ u16 i;
+
+ if (cookie == 0)
+ return -ENXIO;
+
+ /*
+ * Find a complete set of labels by uuid. By definition we can start
+ * with any mapping as the reference label
+ */
+ for_each_label(l, nd_label, nd_region->mapping[0].labels) {
+ u64 isetcookie = readq(&nd_label->isetcookie);
+ u8 label_uuid[NSLABEL_UUID_LEN];
+
+ if (isetcookie != cookie)
+ continue;
+
+ memcpy_fromio(label_uuid, nd_label->uuid,
+ NSLABEL_UUID_LEN);
+ for (i = 0; nd_region->ndr_mappings; i++)
+ if (!has_uuid_at_pos(nd_region, label_uuid, cookie, i))
+ break;
+ if (i < nd_region->ndr_mappings) {
+ /*
+ * Give up if we don't find an instance of a
+ * uuid at each position (from 0 to
+ * nd_region->ndr_mappings - 1), or if we find a
+ * dimm with two instances of the same uuid.
+ */
+ rc = -EINVAL;
+ goto err;
+ } else if (pmem_uuid) {
+ /*
+ * If there is more than one valid uuid set, we
+ * need userspace to clean this up.
+ */
+ rc = -EBUSY;
+ goto err;
+ }
+ memcpy(select_uuid, label_uuid, NSLABEL_UUID_LEN);
+ pmem_uuid = select_uuid;
+ }
+
+ /*
+ * Fix up each mapping's 'labels' to have the validated pmem label for
+ * that position at labels[0], and NULL at labels[1]. In the process,
+ * check that the namespace aligns with interleave-set. We know
+ * that it does not overlap with any blk namespaces by virtue of
+ * the dimm being enabled (i.e. nd_label_reserve_dpa()
+ * succeeded).
+ */
+ rc = select_pmem_uuid(nd_region, pmem_uuid);
+ if (rc)
+ goto err;
+
+ /* Calculate total size and populate namespace properties from label0 */
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ struct nd_namespace_label __iomem *label0;
+
+ label0 = nd_get_label(nd_mapping->labels, 0);
+ size += readq(&label0->rawsize);
+ if (readl(&label0->position) != 0)
+ continue;
+ WARN_ON(nspm->alt_name || nspm->uuid);
+ nspm->alt_name = kmemdup((void __force *) label0->name,
+ NSLABEL_NAME_LEN, GFP_KERNEL);
+ nspm->uuid = kmemdup((void __force *) label0->uuid,
+ NSLABEL_UUID_LEN, GFP_KERNEL);
+ }
+
+ if (!nspm->alt_name || !nspm->uuid) {
+ rc = -ENOMEM;
+ goto err;
+ }
+
+ nd_namespace_pmem_set_size(nd_region, nspm, size);
+
+ return 0;
+ err:
+ switch (rc) {
+ case -EINVAL:
+ dev_dbg(&nd_region->dev, "%s: invalid label(s)\n", __func__);
+ break;
+ case -ENODEV:
+ dev_dbg(&nd_region->dev, "%s: label not found\n", __func__);
+ break;
+ default:
+ dev_dbg(&nd_region->dev, "%s: unexpected err: %d\n", __func__, rc);
+ break;
+ }
+ return rc;
+}
+
+static struct device **create_namespace_pmem(struct nd_region *nd_region)
+{
+ struct nd_namespace_pmem *nspm;
+ struct device *dev, **devs;
+ struct resource *res;
+ int rc;
+
+ nspm = kzalloc(sizeof(*nspm), GFP_KERNEL);
+ if (!nspm)
+ return NULL;
+
+ dev = &nspm->nsio.dev;
+ dev->type = &namespace_pmem_device_type;
+ res = &nspm->nsio.res;
+ res->name = dev_name(&nd_region->dev);
+ res->flags = IORESOURCE_MEM;
+ rc = find_pmem_label_set(nd_region, nspm);
+ if (rc == -ENODEV) {
+ int i;
+
+ /* Pass, try to permit namespace creation... */
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+
+ kfree(nd_mapping->labels);
+ nd_mapping->labels = NULL;
+ }
+
+ /* Publish a zero-sized namespace for userspace to configure. */
+ nd_namespace_pmem_set_size(nd_region, nspm, 0);
+
+ rc = 0;
+ } else if (rc)
+ goto err;
+
+ devs = kcalloc(2, sizeof(struct device *), GFP_KERNEL);
+ if (!devs)
+ goto err;
+
+ devs[0] = dev;
+ return devs;
+
+ err:
+ namespace_pmem_release(&nspm->nsio.dev);
+ return NULL;
+}
+
+static int init_active_labels(struct nd_region *nd_region)
+{
+ int i;
+
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ int count, j;
+
+ /*
+ * If the dimm is disabled then prevent the region from
+ * being activated if it aliases DPA.
+ */
+ if (!ndd) {
+ struct nd_dimm *nd_dimm = nd_mapping->nd_dimm;
+
+ if ((nd_dimm->flags & NDD_ALIASING) == 0)
+ return 0;
+ dev_dbg(&nd_region->dev, "%s: is disabled, failing probe\n",
+ dev_name(&nd_mapping->nd_dimm->dev));
+ return -ENXIO;
+ }
+
+ count = nd_label_active_count(ndd);
+ dev_dbg(ndd->dev, "%s: %d\n", __func__, count);
+ if (!count)
+ continue;
+ nd_mapping->labels = kcalloc(count + 1,
+ sizeof(struct nd_namespace_label *), GFP_KERNEL);
+ if (!nd_mapping->labels)
+ return -ENOMEM;
+ for (j = 0; j < count; j++) {
+ struct nd_namespace_label __iomem *label;
+
+ label = nd_label_active(ndd, j);
+ nd_set_label(nd_mapping->labels, label, j);
+ }
+ }
+
+ return 0;
+}
+
int nd_region_register_namespaces(struct nd_region *nd_region, int *err)
{
struct device **devs = NULL;
- int i;
+ int i, rc = 0, type;
*err = 0;
- switch (nd_region_to_namespace_type(nd_region)) {
+ nd_bus_lock(&nd_region->dev);
+ rc = init_active_labels(nd_region);
+ if (rc) {
+ nd_bus_unlock(&nd_region->dev);
+ return rc;
+ }
+
+ type = nd_region_to_namespace_type(nd_region);
+ switch (type) {
case ND_DEVICE_NAMESPACE_IO:
devs = create_namespace_io(nd_region);
break;
+ case ND_DEVICE_NAMESPACE_PMEM:
+ devs = create_namespace_pmem(nd_region);
+ break;
default:
break;
}
+ nd_bus_unlock(&nd_region->dev);
- if (!devs)
- return -ENODEV;
+ if (!devs) {
+ rc = -ENODEV;
+ goto err;
+ }
+ nd_region->ns_seed = devs[0];
for (i = 0; devs[i]; i++) {
struct device *dev = devs[i];
@@ -108,4 +1114,14 @@ int nd_region_register_namespaces(struct nd_region *nd_region, int *err)
kfree(devs);
return i;
+
+ err:
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+
+ kfree(nd_mapping->labels);
+ nd_mapping->labels = NULL;
+ }
+
+ return rc;
}
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
index 5d8249be3415..f09d7784dc11 100644
--- a/drivers/block/nd/nd-private.h
+++ b/drivers/block/nd/nd-private.h
@@ -60,4 +60,18 @@ int nd_bus_register_dimms(struct nd_bus *nd_bus);
int nd_bus_register_regions(struct nd_bus *nd_bus);
int nd_bus_init_interleave_sets(struct nd_bus *nd_bus);
int nd_match_dimm(struct device *dev, void *data);
+struct nd_label_id;
+char *nd_label_gen_id(struct nd_label_id *label_id, u8 *uuid, u32 flags);
+bool nd_is_uuid_unique(struct device *dev, u8 *uuid);
+struct nd_region;
+struct nd_dimm_drvdata;
+struct nd_mapping;
+resource_size_t nd_pmem_available_dpa(struct nd_region *nd_region,
+ struct nd_mapping *nd_mapping, resource_size_t *overlap);
+resource_size_t nd_region_available_dpa(struct nd_region *nd_region);
+struct resource *nd_dimm_allocate_dpa(struct nd_dimm_drvdata *ndd,
+ struct nd_label_id *label_id, resource_size_t start,
+ resource_size_t n);
+resource_size_t nd_dimm_allocated_dpa(struct nd_dimm_drvdata *ndd,
+ struct nd_label_id *label_id);
#endif /* __ND_PRIVATE_H__ */
diff --git a/drivers/block/nd/nd.h b/drivers/block/nd/nd.h
index 832103a5e3f7..c93675172889 100644
--- a/drivers/block/nd/nd.h
+++ b/drivers/block/nd/nd.h
@@ -15,6 +15,7 @@
#include <linux/device.h>
#include <linux/mutex.h>
#include <linux/ndctl.h>
+#include <linux/types.h>
#include "libnd.h"
#include "label.h"
@@ -59,12 +60,37 @@ static inline struct nd_namespace_index __iomem *to_next_namespace_index(
(unsigned long long) (res ? resource_size(res) : 0), \
(unsigned long long) (res ? res->start : 0), ##arg)
+/* sparse helpers */
+static inline void nd_set_label(struct nd_namespace_label **labels,
+ struct nd_namespace_label __iomem *label, int idx)
+{
+ labels[idx] = (void __force *) label;
+}
+
+static inline struct nd_namespace_label __iomem *nd_get_label(
+ struct nd_namespace_label **labels, int idx)
+{
+ struct nd_namespace_label __iomem *label = NULL;
+
+ if (labels)
+ label = (struct nd_namespace_label __iomem *) labels[idx];
+
+ return label;
+}
+
+#define for_each_label(l, label, labels) \
+ for (l = 0; (label = nd_get_label(labels, l)); l++)
+
+#define for_each_dpa_resource(ndd, res) \
+ for (res = (ndd)->dpa.child; res; res = res->sibling)
+
#define for_each_dpa_resource_safe(ndd, res, next) \
for (res = (ndd)->dpa.child, next = res ? res->sibling : NULL; \
res; res = next, next = next ? next->sibling : NULL)
struct nd_region {
struct device dev;
+ struct device *ns_seed;
u16 ndr_mappings;
u64 ndr_size;
u64 ndr_start;
@@ -88,15 +114,22 @@ enum nd_async_mode {
ND_ASYNC,
};
+void wait_nd_bus_probe_idle(struct device *dev);
void nd_device_register(struct device *dev);
void nd_device_unregister(struct device *dev, enum nd_async_mode mode);
+int nd_uuid_store(struct device *dev, u8 **uuid_out, const char *buf,
+ size_t len);
+struct nd_dimm;
+struct nd_dimm_drvdata *to_ndd(struct nd_mapping *nd_mapping);
int nd_dimm_init_nsarea(struct nd_dimm_drvdata *ndd);
int nd_dimm_init_config_data(struct nd_dimm_drvdata *ndd);
struct nd_region *to_nd_region(struct device *dev);
int nd_region_to_namespace_type(struct nd_region *nd_region);
int nd_region_register_namespaces(struct nd_region *nd_region, int *err);
+u64 nd_region_interleave_set_cookie(struct nd_region *nd_region);
void nd_bus_lock(struct device *dev);
void nd_bus_unlock(struct device *dev);
bool is_nd_bus_locked(struct device *dev);
int nd_label_reserve_dpa(struct nd_dimm_drvdata *ndd);
+void nd_dimm_free_dpa(struct nd_dimm_drvdata *ndd, struct resource *res);
#endif /* __ND_H__ */
diff --git a/drivers/block/nd/pmem.c b/drivers/block/nd/pmem.c
index c9dd8d7eca3a..32e7e74ee25a 100644
--- a/drivers/block/nd/pmem.c
+++ b/drivers/block/nd/pmem.c
@@ -24,6 +24,7 @@
#include <linux/moduleparam.h>
#include <linux/slab.h>
#include <linux/nd.h>
+#include "nd.h"
#define PMEM_MINORS 16
@@ -209,9 +210,27 @@ static void pmem_free(struct pmem_device *pmem)
static int nd_pmem_probe(struct device *dev)
{
+ struct nd_region *nd_region = to_nd_region(dev->parent);
struct nd_namespace_io *nsio = to_nd_namespace_io(dev);
struct pmem_device *pmem;
+ if (resource_size(&nsio->res) < ND_MIN_NAMESPACE_SIZE) {
+ resource_size_t size = resource_size(&nsio->res);
+
+ dev_dbg(dev, "%s: size: %pa, too small must be at least %#x\n",
+ __func__, &size, ND_MIN_NAMESPACE_SIZE);
+ return -ENODEV;
+ }
+
+ if (nd_region_to_namespace_type(nd_region) == ND_DEVICE_NAMESPACE_PMEM) {
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+
+ if (!nspm->uuid) {
+ dev_dbg(dev, "%s: uuid not set\n", __func__);
+ return -ENODEV;
+ }
+ }
+
pmem = pmem_alloc(dev, &nsio->res);
if (IS_ERR(pmem))
return PTR_ERR(pmem);
@@ -231,13 +250,14 @@ static int nd_pmem_remove(struct device *dev)
MODULE_ALIAS("pmem");
MODULE_ALIAS_ND_DEVICE(ND_DEVICE_NAMESPACE_IO);
+MODULE_ALIAS_ND_DEVICE(ND_DEVICE_NAMESPACE_PMEM);
static struct nd_device_driver nd_pmem_driver = {
.probe = nd_pmem_probe,
.remove = nd_pmem_remove,
.drv = {
.name = "pmem",
},
- .type = ND_DRIVER_NAMESPACE_IO,
+ .type = ND_DRIVER_NAMESPACE_IO | ND_DRIVER_NAMESPACE_PMEM,
};
static int __init pmem_init(void)
diff --git a/drivers/block/nd/region_devs.c b/drivers/block/nd/region_devs.c
index c1cea42d0473..e74b3d4534b4 100644
--- a/drivers/block/nd/region_devs.c
+++ b/drivers/block/nd/region_devs.c
@@ -15,6 +15,7 @@
#include <linux/slab.h>
#include <linux/sort.h>
#include <linux/io.h>
+#include <linux/nd.h>
#include "nd-private.h"
#include "nd.h"
@@ -99,6 +100,58 @@ int nd_region_to_namespace_type(struct nd_region *nd_region)
return 0;
}
+EXPORT_SYMBOL(nd_region_to_namespace_type);
+
+static int is_uuid_busy(struct device *dev, void *data)
+{
+ struct nd_region *nd_region = to_nd_region(dev->parent);
+ u8 *uuid = data;
+
+ switch (nd_region_to_namespace_type(nd_region)) {
+ case ND_DEVICE_NAMESPACE_PMEM: {
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+
+ if (!nspm->uuid)
+ break;
+ if (memcmp(uuid, nspm->uuid, NSLABEL_UUID_LEN) == 0)
+ return -EBUSY;
+ break;
+ }
+ case ND_DEVICE_NAMESPACE_BLK: {
+ /* TODO: blk namespace support */
+ break;
+ }
+ default:
+ break;
+ }
+
+ return 0;
+}
+
+static int is_namespace_uuid_busy(struct device *dev, void *data)
+{
+ if (is_nd_pmem(dev) || is_nd_blk(dev))
+ return device_for_each_child(dev, data, is_uuid_busy);
+ return 0;
+}
+
+/**
+ * nd_is_uuid_unique - verify that no other namespace has @uuid
+ * @dev: any device on a nd_bus
+ * @uuid: uuid to check
+ */
+bool nd_is_uuid_unique(struct device *dev, u8 *uuid)
+{
+ struct nd_bus *nd_bus = walk_to_nd_bus(dev);
+
+ if (!nd_bus)
+ return false;
+ WARN_ON_ONCE(!is_nd_bus_locked(&nd_bus->dev));
+ if (device_for_each_child(&nd_bus->dev, uuid,
+ is_namespace_uuid_busy) != 0)
+ return false;
+ return true;
+}
static ssize_t size_show(struct device *dev,
struct device_attribute *attr, char *buf)
@@ -151,6 +204,60 @@ static ssize_t set_cookie_show(struct device *dev,
}
static DEVICE_ATTR_RO(set_cookie);
+resource_size_t nd_region_available_dpa(struct nd_region *nd_region)
+{
+ resource_size_t blk_max_overlap = 0, available, overlap;
+ int i;
+
+ WARN_ON(!is_nd_bus_locked(&nd_region->dev));
+
+ retry:
+ available = 0;
+ overlap = blk_max_overlap;
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+
+ /* if a dimm is disabled the available capacity is zero */
+ if (!ndd)
+ return 0;
+
+ if (is_nd_pmem(&nd_region->dev)) {
+ available += nd_pmem_available_dpa(nd_region,
+ nd_mapping, &overlap);
+ if (overlap > blk_max_overlap) {
+ blk_max_overlap = overlap;
+ goto retry;
+ }
+ } else if (is_nd_blk(&nd_region->dev)) {
+ /* TODO: BLK Namespace support */
+ }
+ }
+
+ return available;
+}
+
+static ssize_t available_size_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_region *nd_region = to_nd_region(dev);
+ unsigned long long available = 0;
+
+ /*
+ * Flush in-flight updates and grab a snapshot of the available
+ * size. Of course, this value is potentially invalidated the
+ * memory nd_bus_lock() is dropped, but that's userspace's
+ * problem to not race itself.
+ */
+ nd_bus_lock(dev);
+ wait_nd_bus_probe_idle(dev);
+ available = nd_region_available_dpa(nd_region);
+ nd_bus_unlock(dev);
+
+ return sprintf(buf, "%llu\n", available);
+}
+static DEVICE_ATTR_RO(available_size);
+
static ssize_t init_namespaces_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
@@ -163,11 +270,29 @@ static ssize_t init_namespaces_show(struct device *dev,
}
static DEVICE_ATTR_RO(init_namespaces);
+static ssize_t namespace_seed_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_region *nd_region = to_nd_region(dev);
+ ssize_t rc;
+
+ nd_bus_lock(dev);
+ if (nd_region->ns_seed)
+ rc = sprintf(buf, "%s\n", dev_name(nd_region->ns_seed));
+ else
+ rc = sprintf(buf, "\n");
+ nd_bus_unlock(dev);
+ return rc;
+}
+static DEVICE_ATTR_RO(namespace_seed);
+
static struct attribute *nd_region_attributes[] = {
&dev_attr_size.attr,
&dev_attr_nstype.attr,
&dev_attr_mappings.attr,
&dev_attr_set_cookie.attr,
+ &dev_attr_available_size.attr,
+ &dev_attr_namespace_seed.attr,
&dev_attr_init_namespaces.attr,
NULL,
};
@@ -177,12 +302,17 @@ static umode_t nd_region_visible(struct kobject *kobj, struct attribute *a, int
struct device *dev = container_of(kobj, typeof(*dev), kobj);
struct nd_region *nd_region = to_nd_region(dev);
struct nd_interleave_set *nd_set = nd_region->nd_set;
+ int type = nd_region_to_namespace_type(nd_region);
- if (a != &dev_attr_set_cookie.attr)
+ if (a != &dev_attr_set_cookie.attr && a != &dev_attr_available_size.attr)
return a->mode;
- if (is_nd_pmem(dev) && nd_set)
- return a->mode;
+ if ((type == ND_DEVICE_NAMESPACE_PMEM
+ || type == ND_DEVICE_NAMESPACE_BLK)
+ && a == &dev_attr_available_size.attr)
+ return a->mode;
+ else if (is_nd_pmem(dev) && nd_set)
+ return a->mode;
return 0;
}
@@ -193,6 +323,15 @@ struct attribute_group nd_region_attribute_group = {
};
EXPORT_SYMBOL_GPL(nd_region_attribute_group);
+u64 nd_region_interleave_set_cookie(struct nd_region *nd_region)
+{
+ struct nd_interleave_set *nd_set = nd_region->nd_set;
+
+ if (nd_set)
+ return nd_set->cookie;
+ return 0;
+}
+
/*
* Upon successful probe/remove, take/release a reference on the
* associated interleave set (if present)
diff --git a/include/linux/nd.h b/include/linux/nd.h
index da70e9962197..255c38a83083 100644
--- a/include/linux/nd.h
+++ b/include/linux/nd.h
@@ -28,16 +28,40 @@ static inline struct nd_device_driver *to_nd_device_driver(
return container_of(drv, struct nd_device_driver, drv);
};
+/**
+ * struct nd_namespace_io - infrastructure for loading an nd_pmem instance
+ * @dev: namespace device created by the nd region driver
+ * @res: struct resource conversion of a NFIT SPA table
+ */
struct nd_namespace_io {
struct device dev;
struct resource res;
};
+/**
+ * struct nd_namespace_pmem - namespace device for dimm-backed interleaved memory
+ * @nsio: device and system physical address range to drive
+ * @alt_name: namespace name supplied in the dimm label
+ * @uuid: namespace name supplied in the dimm label
+ */
+struct nd_namespace_pmem {
+ struct nd_namespace_io nsio;
+ char *alt_name;
+ u8 *uuid;
+};
+
static inline struct nd_namespace_io *to_nd_namespace_io(struct device *dev)
{
return container_of(dev, struct nd_namespace_io, dev);
}
+static inline struct nd_namespace_pmem *to_nd_namespace_pmem(struct device *dev)
+{
+ struct nd_namespace_io *nsio = to_nd_namespace_io(dev);
+
+ return container_of(nsio, struct nd_namespace_pmem, nsio);
+}
+
#define MODULE_ALIAS_ND_DEVICE(type) \
MODULE_ALIAS("nd:t" __stringify(type) "*")
#define ND_DEVICE_MODALIAS_FMT "nd:t%d"
diff --git a/include/uapi/linux/ndctl.h b/include/uapi/linux/ndctl.h
index 624a19d9e6e4..0b4dcabb248a 100644
--- a/include/uapi/linux/ndctl.h
+++ b/include/uapi/linux/ndctl.h
@@ -190,4 +190,8 @@ enum nd_driver_flags {
ND_DRIVER_NAMESPACE_PMEM = 1 << ND_DEVICE_NAMESPACE_PMEM,
ND_DRIVER_NAMESPACE_BLK = 1 << ND_DEVICE_NAMESPACE_BLK,
};
+
+enum {
+ ND_MIN_NAMESPACE_SIZE = 0x00400000,
+};
#endif /* __NDCTL_H__ */
A blk label set describes a namespace comprised of one or more
discontiguous dpa ranges on a single dimm. They may alias with one or
more pmem interleave sets that include the given dimm.
This is the runtime/volatile configuration infrastructure for sysfs
manipulation of 'alt_name', 'uuid', 'size', and 'sector_size'. A later
patch will make these settings persistent by writing back the label(s).
Unlike pmem namespaces, multiple blk namespaces can be created per
region. Once a blk namespace has been created a new seed device
(unconfigured child of a parent blk region) is instantiated. As long as
a region has 'available_size' != 0 new child namespaces may be created.
Cc: Greg KH <[email protected]>
Cc: Neil Brown <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/core.c | 40 +++
drivers/block/nd/dimm_devs.c | 35 +++
drivers/block/nd/libnd.h | 3
drivers/block/nd/namespace_devs.c | 502 ++++++++++++++++++++++++++++++++++---
drivers/block/nd/nd-private.h | 8 +
drivers/block/nd/nd.h | 5
drivers/block/nd/region_devs.c | 15 +
include/linux/nd.h | 25 ++
8 files changed, 589 insertions(+), 44 deletions(-)
diff --git a/drivers/block/nd/core.c b/drivers/block/nd/core.c
index cf64b7a50d3a..3ec38289be58 100644
--- a/drivers/block/nd/core.c
+++ b/drivers/block/nd/core.c
@@ -170,6 +170,46 @@ int nd_uuid_store(struct device *dev, u8 **uuid_out, const char *buf,
return 0;
}
+ssize_t nd_sector_size_show(unsigned long current_lbasize,
+ const unsigned long *supported, char *buf)
+{
+ ssize_t len = 0;
+ int i;
+
+ for (i = 0; supported[i]; i++)
+ if (current_lbasize == supported[i])
+ len += sprintf(buf + len, "[%ld] ", supported[i]);
+ else
+ len += sprintf(buf + len, "%ld ", supported[i]);
+ len += sprintf(buf + len, "\n");
+ return len;
+}
+
+ssize_t nd_sector_size_store(struct device *dev, const char *buf,
+ unsigned long *current_lbasize, const unsigned long *supported)
+{
+ unsigned long lbasize;
+ int rc, i;
+
+ if (dev->driver)
+ return -EBUSY;
+
+ rc = kstrtoul(buf, 0, &lbasize);
+ if (rc)
+ return rc;
+
+ for (i = 0; supported[i]; i++)
+ if (lbasize == supported[i])
+ break;
+
+ if (supported[i]) {
+ *current_lbasize = lbasize;
+ return 0;
+ } else {
+ return -EINVAL;
+ }
+}
+
static ssize_t commands_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
diff --git a/drivers/block/nd/dimm_devs.c b/drivers/block/nd/dimm_devs.c
index b242d3ae6d12..4aa5654354ac 100644
--- a/drivers/block/nd/dimm_devs.c
+++ b/drivers/block/nd/dimm_devs.c
@@ -256,6 +256,41 @@ struct nd_dimm *nd_dimm_create(struct nd_bus *nd_bus, void *provider_data,
EXPORT_SYMBOL_GPL(nd_dimm_create);
/**
+ * nd_blk_available_dpa - account the unused dpa of BLK region
+ * @nd_mapping: container of dpa-resource-root + labels
+ *
+ * Unlike PMEM, BLK namespaces can occupy discontiguous DPA ranges.
+ */
+resource_size_t nd_blk_available_dpa(struct nd_mapping *nd_mapping)
+{
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ resource_size_t map_end, busy = 0, available;
+ struct resource *res;
+
+ if (!ndd)
+ return 0;
+
+ map_end = nd_mapping->start + nd_mapping->size - 1;
+ for_each_dpa_resource(ndd, res)
+ if (res->start >= nd_mapping->start && res->start < map_end) {
+ resource_size_t end = min(map_end, res->end);
+
+ busy += end - res->start + 1;
+ } else if (res->end >= nd_mapping->start && res->end <= map_end) {
+ busy += res->end - nd_mapping->start;
+ } else if (nd_mapping->start > res->start
+ && nd_mapping->start < res->end) {
+ /* total eclipse of the BLK region mapping */
+ busy += nd_mapping->size;
+ }
+
+ available = map_end - nd_mapping->start + 1;
+ if (busy < available)
+ return available - busy;
+ return 0;
+}
+
+/**
* nd_pmem_available_dpa - for the given dimm+region account unallocated dpa
* @nd_mapping: container of dpa-resource-root + labels
* @nd_region: constrain available space check to this reference region
diff --git a/drivers/block/nd/libnd.h b/drivers/block/nd/libnd.h
index 832dcfebbb49..3f6b5e09cd67 100644
--- a/drivers/block/nd/libnd.h
+++ b/drivers/block/nd/libnd.h
@@ -26,6 +26,9 @@ enum {
ND_CMD_MAX_ENVELOPE = 16,
ND_CMD_ARS_QUERY_MAX = SZ_4K,
ND_MAX_MAPPINGS = 32,
+
+ /* mark newly adjusted resources as requiring a label update */
+ DPA_RESOURCE_ADJUSTED = 1 << 0,
};
extern struct attribute_group nd_bus_attribute_group;
diff --git a/drivers/block/nd/namespace_devs.c b/drivers/block/nd/namespace_devs.c
index fb23904df4c7..e229755509a2 100644
--- a/drivers/block/nd/namespace_devs.c
+++ b/drivers/block/nd/namespace_devs.c
@@ -37,7 +37,15 @@ static void namespace_pmem_release(struct device *dev)
static void namespace_blk_release(struct device *dev)
{
- /* TODO: blk namespace support */
+ struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+ struct nd_region *nd_region = to_nd_region(dev->parent);
+
+ if (nsblk->id >= 0)
+ ida_simple_remove(&nd_region->ns_ida, nsblk->id);
+ kfree(nsblk->alt_name);
+ kfree(nsblk->uuid);
+ kfree(nsblk->res);
+ kfree(nsblk);
}
static struct device_type namespace_io_device_type = {
@@ -90,8 +98,9 @@ static ssize_t __alt_name_store(struct device *dev, const char *buf,
ns_altname = &nspm->alt_name;
} else if (is_namespace_blk(dev)) {
- /* TODO: blk namespace support */
- return -ENXIO;
+ struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+
+ ns_altname = &nsblk->alt_name;
} else
return -ENXIO;
@@ -124,6 +133,24 @@ out:
return rc;
}
+static resource_size_t nd_namespace_blk_size(struct nd_namespace_blk *nsblk)
+{
+ struct nd_region *nd_region = to_nd_region(nsblk->dev.parent);
+ struct nd_mapping *nd_mapping = &nd_region->mapping[0];
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ struct nd_label_id label_id;
+ resource_size_t size = 0;
+ struct resource *res;
+
+ if (!nsblk->uuid)
+ return 0;
+ nd_label_gen_id(&label_id, nsblk->uuid, NSLABEL_FLAG_LOCAL);
+ for_each_dpa_resource(ndd, res)
+ if (strcmp(res->name, label_id.id) == 0)
+ size += resource_size(res);
+ return size;
+}
+
static ssize_t alt_name_store(struct device *dev,
struct device_attribute *attr, const char *buf, size_t len)
{
@@ -150,8 +177,9 @@ static ssize_t alt_name_show(struct device *dev,
ns_altname = nspm->alt_name;
} else if (is_namespace_blk(dev)) {
- /* TODO: blk namespace support */
- return -ENXIO;
+ struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+
+ ns_altname = nsblk->alt_name;
} else
return -ENXIO;
@@ -197,6 +225,8 @@ static int scan_free(struct nd_region *nd_region,
new_start = res->start;
rc = adjust_resource(res, new_start, resource_size(res) - n);
+ if (rc == 0)
+ res->flags |= DPA_RESOURCE_ADJUSTED;
nd_dbg_dpa(nd_region, ndd, res, "shrink %d\n", rc);
break;
}
@@ -257,14 +287,15 @@ static resource_size_t init_dpa_allocation(struct nd_label_id *label_id,
return rc ? n : 0;
}
-static bool space_valid(bool is_pmem, struct nd_label_id *label_id,
- struct resource *res)
+static bool space_valid(bool is_pmem, bool is_reserve,
+ struct nd_label_id *label_id, struct resource *res)
{
/*
* For BLK-space any space is valid, for PMEM-space, it must be
- * contiguous with an existing allocation.
+ * contiguous with an existing allocation unless we are
+ * reserving pmem.
*/
- if (!is_pmem)
+ if (is_reserve || !is_pmem)
return true;
if (!res || strcmp(res->name, label_id->id) == 0)
return true;
@@ -280,6 +311,7 @@ static resource_size_t scan_allocate(struct nd_region *nd_region,
resource_size_t n)
{
resource_size_t mapping_end = nd_mapping->start + nd_mapping->size - 1;
+ bool is_reserve = strcmp(label_id->id, "pmem-reserve") == 0;
bool is_pmem = strncmp(label_id->id, "pmem", 4) == 0;
struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
const resource_size_t to_allocate = n;
@@ -305,7 +337,7 @@ static resource_size_t scan_allocate(struct nd_region *nd_region,
if (!first++ && res->start > nd_mapping->start) {
free_start = nd_mapping->start;
available = res->start - free_start;
- if (space_valid(is_pmem, label_id, NULL))
+ if (space_valid(is_pmem, is_reserve, label_id, NULL))
loc = ALLOC_BEFORE;
}
@@ -313,7 +345,7 @@ static resource_size_t scan_allocate(struct nd_region *nd_region,
if (!loc && next) {
free_start = res->start + resource_size(res);
free_end = min(mapping_end, next->start - 1);
- if (space_valid(is_pmem, label_id, res)
+ if (space_valid(is_pmem, is_reserve, label_id, res)
&& free_start < free_end) {
available = free_end + 1 - free_start;
loc = ALLOC_MID;
@@ -324,7 +356,7 @@ static resource_size_t scan_allocate(struct nd_region *nd_region,
if (!loc && !next) {
free_start = res->start + resource_size(res);
free_end = mapping_end;
- if (space_valid(is_pmem, label_id, res)
+ if (space_valid(is_pmem, is_reserve, label_id, res)
&& free_start < free_end) {
available = free_end + 1 - free_start;
loc = ALLOC_AFTER;
@@ -338,7 +370,7 @@ static resource_size_t scan_allocate(struct nd_region *nd_region,
case ALLOC_BEFORE:
if (strcmp(res->name, label_id->id) == 0) {
/* adjust current resource up */
- if (is_pmem)
+ if (is_pmem && !is_reserve)
return n;
rc = adjust_resource(res, res->start - allocate,
resource_size(res) + allocate);
@@ -349,7 +381,7 @@ static resource_size_t scan_allocate(struct nd_region *nd_region,
case ALLOC_MID:
if (strcmp(next->name, label_id->id) == 0) {
/* adjust next resource up */
- if (is_pmem)
+ if (is_pmem && !is_reserve)
return n;
rc = adjust_resource(next, next->start
- allocate, resource_size(next)
@@ -375,7 +407,7 @@ static resource_size_t scan_allocate(struct nd_region *nd_region,
/* BLK allocate bottom up */
if (!is_pmem)
free_start += available - allocate;
- else if (free_start != nd_mapping->start)
+ else if (!is_reserve && free_start != nd_mapping->start)
return n;
new_res = nd_dimm_allocate_dpa(ndd, label_id,
@@ -386,6 +418,8 @@ static resource_size_t scan_allocate(struct nd_region *nd_region,
/* adjust current resource down */
rc = adjust_resource(res, res->start, resource_size(res)
+ allocate);
+ if (rc == 0)
+ res->flags |= DPA_RESOURCE_ADJUSTED;
}
if (!new_res)
@@ -411,11 +445,106 @@ static resource_size_t scan_allocate(struct nd_region *nd_region,
return 0;
}
- if (is_pmem && n == to_allocate)
+ /*
+ * If we allocated nothing in the BLK case it may be because we are in
+ * an initial "pmem-reserve pass". Only do an initial BLK allocation
+ * when none of the DPA space is reserved.
+ */
+ if ((is_pmem || !ndd->dpa.child) && n == to_allocate)
return init_dpa_allocation(label_id, nd_region, nd_mapping, n);
return n;
}
+static int merge_dpa(struct nd_region *nd_region,
+ struct nd_mapping *nd_mapping, struct nd_label_id *label_id)
+{
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ struct resource *res;
+
+ if (strncmp("pmem", label_id->id, 4) == 0)
+ return 0;
+ retry:
+ for_each_dpa_resource(ndd, res) {
+ int rc;
+ struct resource *next = res->sibling;
+ resource_size_t end = res->start + resource_size(res);
+
+ if (!next || strcmp(res->name, label_id->id) != 0
+ || strcmp(next->name, label_id->id) != 0
+ || end != next->start)
+ continue;
+ end += resource_size(next);
+ nd_dimm_free_dpa(ndd, next);
+ rc = adjust_resource(res, res->start, end - res->start);
+ nd_dbg_dpa(nd_region, ndd, res, "merge %d\n", rc);
+ if (rc)
+ return rc;
+ res->flags |= DPA_RESOURCE_ADJUSTED;
+ goto retry;
+ }
+
+ return 0;
+}
+
+static int __reserve_free_pmem(struct device *dev, void *data)
+{
+ struct nd_dimm *nd_dimm = data;
+ struct nd_region *nd_region;
+ struct nd_label_id label_id;
+ int i;
+
+ if (!is_nd_pmem(dev))
+ return 0;
+
+ nd_region = to_nd_region(dev);
+ if (nd_region->ndr_mappings == 0)
+ return 0;
+
+ memset(&label_id, 0, sizeof(label_id));
+ strcat(label_id.id, "pmem-reserve");
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ resource_size_t n, rem = 0;
+
+ if (nd_mapping->nd_dimm != nd_dimm)
+ continue;
+
+ n = nd_pmem_available_dpa(nd_region, nd_mapping, &rem);
+ if (n == 0)
+ return 0;
+ rem = scan_allocate(nd_region, nd_mapping, &label_id, n);
+ dev_WARN_ONCE(&nd_region->dev, rem,
+ "pmem reserve underrun: %#llx of %#llx bytes\n",
+ (unsigned long long) n - rem,
+ (unsigned long long) n);
+ return rem ? -ENXIO : 0;
+ }
+
+ return 0;
+}
+
+static void release_free_pmem(struct nd_bus *nd_bus, struct nd_mapping *nd_mapping)
+{
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ struct resource *res, *_res;
+
+ for_each_dpa_resource_safe(ndd, res, _res)
+ if (strcmp(res->name, "pmem-reserve") == 0)
+ nd_dimm_free_dpa(ndd, res);
+}
+
+static int reserve_free_pmem(struct nd_bus *nd_bus,
+ struct nd_mapping *nd_mapping)
+{
+ struct nd_dimm *nd_dimm = nd_mapping->nd_dimm;
+ int rc;
+
+ rc = device_for_each_child(&nd_bus->dev, nd_dimm, __reserve_free_pmem);
+ if (rc)
+ release_free_pmem(nd_bus, nd_mapping);
+ return rc;
+}
+
/**
* grow_dpa_allocation - for each dimm allocate n bytes for @label_id
* @nd_region: the set of dimms to allocate @n more bytes from
@@ -432,13 +561,44 @@ static resource_size_t scan_allocate(struct nd_region *nd_region,
static int grow_dpa_allocation(struct nd_region *nd_region,
struct nd_label_id *label_id, resource_size_t n)
{
+ struct nd_bus *nd_bus = walk_to_nd_bus(&nd_region->dev);
+ bool is_pmem = strncmp(label_id->id, "pmem", 4) == 0;
int i;
for (i = 0; i < nd_region->ndr_mappings; i++) {
struct nd_mapping *nd_mapping = &nd_region->mapping[i];
- int rc;
+ resource_size_t rem = n;
+ int rc, j;
- rc = scan_allocate(nd_region, nd_mapping, label_id, n);
+ /*
+ * In the BLK case try once with all unallocated PMEM
+ * reserved, and once without
+ */
+ for (j = is_pmem; j < 2; j++) {
+ bool blk_only = j == 0;
+
+ if (blk_only) {
+ rc = reserve_free_pmem(nd_bus, nd_mapping);
+ if (rc)
+ return rc;
+ }
+ rem = scan_allocate(nd_region, nd_mapping, label_id, rem);
+ if (blk_only)
+ release_free_pmem(nd_bus, nd_mapping);
+
+ /* try again and allow encroachments into PMEM */
+ if (rem == 0)
+ break;
+ }
+
+ dev_WARN_ONCE(&nd_region->dev, rem,
+ "allocation underrun: %#llx of %#llx bytes\n",
+ (unsigned long long) n - rem,
+ (unsigned long long) n);
+ if (rem)
+ return -ENXIO;
+
+ rc = merge_dpa(nd_region, nd_mapping, label_id);
if (rc)
return rc;
}
@@ -474,8 +634,10 @@ static ssize_t __size_store(struct device *dev, unsigned long long val)
uuid = nspm->uuid;
} else if (is_namespace_blk(dev)) {
- /* TODO: blk namespace support */
- return -ENXIO;
+ struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+
+ uuid = nsblk->uuid;
+ flags = NSLABEL_FLAG_LOCAL;
}
/*
@@ -529,6 +691,14 @@ static ssize_t __size_store(struct device *dev, unsigned long long val)
nd_namespace_pmem_set_size(nd_region, nspm,
val * nd_region->ndr_mappings);
+ } else if (is_namespace_blk(dev)) {
+ /*
+ * Try to delete the namespace if we deleted all of its
+ * allocation and this is not the seed device for the
+ * region.
+ */
+ if (val == 0 && nd_region->ns_seed != dev)
+ nd_device_unregister(dev, ND_ASYNC);
}
return rc;
@@ -555,8 +725,9 @@ static ssize_t size_store(struct device *dev,
uuid = &nspm->uuid;
} else if (is_namespace_blk(dev)) {
- /* TODO: blk namespace support */
- rc = -ENXIO;
+ struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+
+ uuid = &nsblk->uuid;
}
if (rc == 0 && val == 0 && uuid) {
@@ -577,21 +748,23 @@ static ssize_t size_store(struct device *dev,
static ssize_t size_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
+ unsigned long long size = 0;
+
+ nd_bus_lock(dev);
if (is_namespace_pmem(dev)) {
struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
- return sprintf(buf, "%llu\n", (unsigned long long)
- resource_size(&nspm->nsio.res));
+ size = resource_size(&nspm->nsio.res);
} else if (is_namespace_blk(dev)) {
- /* TODO: blk namespace support */
- return -ENXIO;
+ size = nd_namespace_blk_size(to_nd_namespace_blk(dev));
} else if (is_namespace_io(dev)) {
struct nd_namespace_io *nsio = to_nd_namespace_io(dev);
- return sprintf(buf, "%llu\n", (unsigned long long)
- resource_size(&nsio->res));
- } else
- return -ENXIO;
+ size = resource_size(&nsio->res);
+ }
+ nd_bus_unlock(dev);
+
+ return sprintf(buf, "%llu\n", size);
}
static DEVICE_ATTR(size, S_IRUGO, size_show, size_store);
@@ -605,8 +778,9 @@ static ssize_t uuid_show(struct device *dev,
uuid = nspm->uuid;
} else if (is_namespace_blk(dev)) {
- /* TODO: blk namespace support */
- return -ENXIO;
+ struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+
+ uuid = nsblk->uuid;
} else
return -ENXIO;
@@ -670,8 +844,9 @@ static ssize_t uuid_store(struct device *dev,
ns_uuid = &nspm->uuid;
} else if (is_namespace_blk(dev)) {
- /* TODO: blk namespace support */
- return -ENXIO;
+ struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+
+ ns_uuid = &nsblk->uuid;
} else
return -ENXIO;
@@ -713,12 +888,48 @@ static ssize_t resource_show(struct device *dev,
}
static DEVICE_ATTR_RO(resource);
+static const unsigned long ns_lbasize_supported[] = { 512, 0 };
+
+static ssize_t sector_size_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+
+ if (!is_namespace_blk(dev))
+ return -ENXIO;
+
+ return nd_sector_size_show(nsblk->lbasize, ns_lbasize_supported, buf);
+}
+
+static ssize_t sector_size_store(struct device *dev,
+ struct device_attribute *attr, const char *buf, size_t len)
+{
+ struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+ ssize_t rc;
+
+ if (!is_namespace_blk(dev))
+ return -ENXIO;
+
+ device_lock(dev);
+ nd_bus_lock(dev);
+ rc = nd_sector_size_store(dev, buf, &nsblk->lbasize,
+ ns_lbasize_supported);
+ dev_dbg(dev, "%s: result: %zd wrote: %s%s", __func__,
+ rc, buf, buf[len - 1] == '\n' ? "" : "\n");
+ nd_bus_unlock(dev);
+ device_unlock(dev);
+
+ return rc ? rc : len;
+}
+static DEVICE_ATTR_RW(sector_size);
+
static struct attribute *nd_namespace_attributes[] = {
&dev_attr_type.attr,
&dev_attr_size.attr,
&dev_attr_uuid.attr,
&dev_attr_resource.attr,
&dev_attr_alt_name.attr,
+ &dev_attr_sector_size.attr,
NULL,
};
@@ -735,6 +946,10 @@ static umode_t nd_namespace_attr_visible(struct kobject *kobj, struct attribute
if (is_namespace_pmem(dev) || is_namespace_blk(dev)) {
if (a == &dev_attr_size.attr)
return S_IWUSR;
+
+ if (is_namespace_pmem(dev) && a == &dev_attr_sector_size.attr)
+ return 0;
+
return a->mode;
}
@@ -1029,6 +1244,173 @@ static struct device **create_namespace_pmem(struct nd_region *nd_region)
return NULL;
}
+struct resource *nsblk_add_resource(struct nd_region *nd_region,
+ struct nd_dimm_drvdata *ndd, struct nd_namespace_blk *nsblk,
+ resource_size_t start)
+{
+ struct nd_label_id label_id;
+ struct resource *res;
+
+ nd_label_gen_id(&label_id, nsblk->uuid, NSLABEL_FLAG_LOCAL);
+ nsblk->res = krealloc(nsblk->res,
+ sizeof(void *) * (nsblk->num_resources + 1),
+ GFP_KERNEL);
+ if (!nsblk->res)
+ return NULL;
+ for_each_dpa_resource(ndd, res)
+ if (strcmp(res->name, label_id.id) == 0 && res->start == start) {
+ nsblk->res[nsblk->num_resources++] = res;
+ return res;
+ }
+ return NULL;
+}
+
+static struct device *nd_namespace_blk_create(struct nd_region *nd_region)
+{
+ struct nd_namespace_blk *nsblk;
+ struct device *dev;
+
+ if (!is_nd_blk(&nd_region->dev))
+ return NULL;
+
+ nsblk = kzalloc(sizeof(*nsblk), GFP_KERNEL);
+ if (!nsblk)
+ return NULL;
+
+ dev = &nsblk->dev;
+ dev->type = &namespace_blk_device_type;
+ nsblk->id = ida_simple_get(&nd_region->ns_ida, 0, 0, GFP_KERNEL);
+ if (nsblk->id < 0) {
+ kfree(nsblk);
+ return NULL;
+ }
+ dev_set_name(dev, "namespace%d.%d", nd_region->id, nsblk->id);
+ dev->parent = &nd_region->dev;
+ dev->groups = nd_namespace_attribute_groups;
+
+ return &nsblk->dev;
+}
+
+void nd_region_create_blk_seed(struct nd_region *nd_region)
+{
+ WARN_ON(!is_nd_bus_locked(&nd_region->dev));
+ nd_region->ns_seed = nd_namespace_blk_create(nd_region);
+ /*
+ * Seed creation failures are not fatal, provisioning is simply
+ * disabled until memory becomes available
+ */
+ if (!nd_region->ns_seed)
+ dev_err(&nd_region->dev, "failed to create blk namespace\n");
+ else
+ nd_device_register(nd_region->ns_seed);
+}
+
+static struct device **create_namespace_blk(struct nd_region *nd_region)
+{
+ struct nd_mapping *nd_mapping = &nd_region->mapping[0];
+ struct nd_namespace_label __iomem *nd_label;
+ struct device *dev, **devs = NULL;
+ u8 label_uuid[NSLABEL_UUID_LEN];
+ struct nd_namespace_blk *nsblk;
+ struct nd_dimm_drvdata *ndd;
+ int i, l, count = 0;
+ struct resource *res;
+
+ if (nd_region->ndr_mappings == 0)
+ return NULL;
+
+ ndd = to_ndd(nd_mapping);
+ for_each_label(l, nd_label, nd_mapping->labels) {
+ u32 flags = readl(&nd_label->flags);
+ char *name[NSLABEL_NAME_LEN];
+ struct device **__devs;
+
+ if (flags & NSLABEL_FLAG_LOCAL)
+ /* pass */;
+ else
+ continue;
+
+ memcpy_fromio(label_uuid, nd_label->uuid, NSLABEL_UUID_LEN);
+ for (i = 0; i < count; i++) {
+ nsblk = to_nd_namespace_blk(devs[i]);
+ if (memcmp(nsblk->uuid, label_uuid,
+ NSLABEL_UUID_LEN) == 0) {
+ res = nsblk_add_resource(nd_region, ndd, nsblk,
+ readq(&nd_label->dpa));
+ if (!res)
+ goto err;
+ nd_dbg_dpa(nd_region, ndd, res, "%s assign\n",
+ dev_name(&nsblk->dev));
+ break;
+ }
+ }
+ if (i < count)
+ continue;
+ __devs = kcalloc(count + 2, sizeof(dev), GFP_KERNEL);
+ if (!__devs)
+ goto err;
+ memcpy(__devs, devs, sizeof(dev) * count);
+ kfree(devs);
+ devs = __devs;
+
+ nsblk = kzalloc(sizeof(*nsblk), GFP_KERNEL);
+ if (!nsblk)
+ goto err;
+ dev = &nsblk->dev;
+ dev->type = &namespace_blk_device_type;
+ dev_set_name(dev, "namespace%d.%d", nd_region->id, count);
+ devs[count++] = dev;
+ nsblk->id = -1;
+ nsblk->lbasize = readq(&nd_label->lbasize);
+ nsblk->uuid = kmemdup(label_uuid, NSLABEL_UUID_LEN, GFP_KERNEL);
+ if (!nsblk->uuid)
+ goto err;
+ memcpy_fromio(name, nd_label->name, NSLABEL_NAME_LEN);
+ if (name[0])
+ nsblk->alt_name = kmemdup(name, NSLABEL_NAME_LEN,
+ GFP_KERNEL);
+ res = nsblk_add_resource(nd_region, ndd, nsblk,
+ readq(&nd_label->dpa));
+ if (!res)
+ goto err;
+ nd_dbg_dpa(nd_region, ndd, res, "%s assign\n",
+ dev_name(&nsblk->dev));
+ }
+
+ dev_dbg(&nd_region->dev, "%s: discovered %d blk namespace%s\n",
+ __func__, count, count == 1 ? "" : "s");
+
+ if (count == 0) {
+ /* Publish a zero-sized namespace for userspace to configure. */
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+
+ kfree(nd_mapping->labels);
+ nd_mapping->labels = NULL;
+ }
+
+ devs = kcalloc(2, sizeof(dev), GFP_KERNEL);
+ if (!devs)
+ goto err;
+ nsblk = kzalloc(sizeof(*nsblk), GFP_KERNEL);
+ if (!nsblk)
+ goto err;
+ dev = &nsblk->dev;
+ dev->type = &namespace_blk_device_type;
+ devs[count++] = dev;
+ }
+
+ return devs;
+
+err:
+ for (i = 0; i < count; i++) {
+ nsblk = to_nd_namespace_blk(devs[i]);
+ namespace_blk_release(&nsblk->dev);
+ }
+ kfree(devs);
+ return NULL;
+}
+
static int init_active_labels(struct nd_region *nd_region)
{
int i;
@@ -1092,6 +1474,9 @@ int nd_region_register_namespaces(struct nd_region *nd_region, int *err)
case ND_DEVICE_NAMESPACE_PMEM:
devs = create_namespace_pmem(nd_region);
break;
+ case ND_DEVICE_NAMESPACE_BLK:
+ devs = create_namespace_blk(nd_region);
+ break;
default:
break;
}
@@ -1105,23 +1490,56 @@ int nd_region_register_namespaces(struct nd_region *nd_region, int *err)
nd_region->ns_seed = devs[0];
for (i = 0; devs[i]; i++) {
struct device *dev = devs[i];
+ int id;
+
+ if (type == ND_DEVICE_NAMESPACE_BLK) {
+ struct nd_namespace_blk *nsblk;
- dev_set_name(dev, "namespace%d.%d", nd_region->id, i);
+ nsblk = to_nd_namespace_blk(dev);
+ id = ida_simple_get(&nd_region->ns_ida, 0, 0,
+ GFP_KERNEL);
+ nsblk->id = id;
+ } else
+ id = i;
+
+ if (id < 0)
+ break;
+ dev_set_name(dev, "namespace%d.%d", nd_region->id, id);
dev->parent = &nd_region->dev;
dev->groups = nd_namespace_attribute_groups;
nd_device_register(dev);
}
- kfree(devs);
- return i;
+ if (devs[i]) {
+ int j;
+
+ for (j = i; devs[j]; j++) {
+ struct device *dev = devs[j];
+
+ device_initialize(dev);
+ put_device(dev);
+ }
+ *err = j - i;
+ /*
+ * All of the namespaces we tried to register failed, so
+ * fail region activation.
+ */
+ if (*err == 0)
+ rc = -ENODEV;
+ }
+ kfree(devs);
err:
- for (i = 0; i < nd_region->ndr_mappings; i++) {
- struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ if (rc == -ENODEV) {
+ nd_region->ns_seed = NULL;
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
- kfree(nd_mapping->labels);
- nd_mapping->labels = NULL;
+ kfree(nd_mapping->labels);
+ nd_mapping->labels = NULL;
+ }
+ return rc;
}
- return rc;
+ return i;
}
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
index f09d7784dc11..43dc6dd36b42 100644
--- a/drivers/block/nd/nd-private.h
+++ b/drivers/block/nd/nd-private.h
@@ -15,6 +15,7 @@
#include <linux/device.h>
#include <linux/sizes.h>
#include <linux/mutex.h>
+#include <linux/nd.h>
#include "libnd.h"
extern struct list_head nd_bus_list;
@@ -52,6 +53,8 @@ void nd_dimm_exit(void);
int nd_region_exit(void);
void nd_region_probe_start(struct nd_bus *nd_bus, struct device *dev);
void nd_region_probe_end(struct nd_bus *nd_bus, struct device *dev, int rc);
+struct nd_region;
+void nd_region_create_blk_seed(struct nd_region *nd_region);
void nd_region_notify_remove(struct nd_bus *nd_bus, struct device *dev, int rc);
int nd_bus_create_ndctl(struct nd_bus *nd_bus);
void nd_bus_destroy_ndctl(struct nd_bus *nd_bus);
@@ -68,10 +71,15 @@ struct nd_dimm_drvdata;
struct nd_mapping;
resource_size_t nd_pmem_available_dpa(struct nd_region *nd_region,
struct nd_mapping *nd_mapping, resource_size_t *overlap);
+resource_size_t nd_blk_available_dpa(struct nd_mapping *nd_mapping);
resource_size_t nd_region_available_dpa(struct nd_region *nd_region);
struct resource *nd_dimm_allocate_dpa(struct nd_dimm_drvdata *ndd,
struct nd_label_id *label_id, resource_size_t start,
resource_size_t n);
resource_size_t nd_dimm_allocated_dpa(struct nd_dimm_drvdata *ndd,
struct nd_label_id *label_id);
+struct nd_mapping;
+struct resource *nsblk_add_resource(struct nd_region *nd_region,
+ struct nd_dimm_drvdata *ndd, struct nd_namespace_blk *nsblk,
+ resource_size_t start);
#endif /* __ND_PRIVATE_H__ */
diff --git a/drivers/block/nd/nd.h b/drivers/block/nd/nd.h
index c93675172889..63d4c6c09275 100644
--- a/drivers/block/nd/nd.h
+++ b/drivers/block/nd/nd.h
@@ -90,6 +90,7 @@ static inline struct nd_namespace_label __iomem *nd_get_label(
struct nd_region {
struct device dev;
+ struct ida ns_ida;
struct device *ns_seed;
u16 ndr_mappings;
u64 ndr_size;
@@ -119,6 +120,10 @@ void nd_device_register(struct device *dev);
void nd_device_unregister(struct device *dev, enum nd_async_mode mode);
int nd_uuid_store(struct device *dev, u8 **uuid_out, const char *buf,
size_t len);
+ssize_t nd_sector_size_show(unsigned long current_lbasize,
+ const unsigned long *supported, char *buf);
+ssize_t nd_sector_size_store(struct device *dev, const char *buf,
+ unsigned long *current_lbasize, const unsigned long *supported);
struct nd_dimm;
struct nd_dimm_drvdata *to_ndd(struct nd_mapping *nd_mapping);
int nd_dimm_init_nsarea(struct nd_dimm_drvdata *ndd);
diff --git a/drivers/block/nd/region_devs.c b/drivers/block/nd/region_devs.c
index e74b3d4534b4..bcdd8e1e21a2 100644
--- a/drivers/block/nd/region_devs.c
+++ b/drivers/block/nd/region_devs.c
@@ -118,7 +118,12 @@ static int is_uuid_busy(struct device *dev, void *data)
break;
}
case ND_DEVICE_NAMESPACE_BLK: {
- /* TODO: blk namespace support */
+ struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+
+ if (!nsblk->uuid)
+ break;
+ if (memcmp(uuid, nsblk->uuid, NSLABEL_UUID_LEN) == 0)
+ return -EBUSY;
break;
}
default:
@@ -230,7 +235,7 @@ resource_size_t nd_region_available_dpa(struct nd_region *nd_region)
goto retry;
}
} else if (is_nd_blk(&nd_region->dev)) {
- /* TODO: BLK Namespace support */
+ available += nd_blk_available_dpa(nd_mapping);
}
}
@@ -355,6 +360,11 @@ static void nd_region_notify_driver_action(struct nd_bus *nd_bus,
else
atomic_dec(&nd_dimm->busy);
}
+ } else if (dev->parent && is_nd_blk(dev->parent) && probe && rc == 0) {
+ struct nd_region *nd_region = to_nd_region(dev->parent);
+
+ if (nd_region->ns_seed == dev)
+ nd_region_create_blk_seed(nd_region);
}
}
@@ -541,6 +551,7 @@ static noinline struct nd_region *nd_region_create(struct nd_bus *nd_bus,
nd_region->ndr_mappings = ndr_desc->num_mappings;
nd_region->provider_data = ndr_desc->provider_data;
nd_region->nd_set = ndr_desc->nd_set;
+ ida_init(&nd_region->ns_ida);
dev = &nd_region->dev;
dev_set_name(dev, "region%d", nd_region->id);
dev->parent = &nd_bus->dev;
diff --git a/include/linux/nd.h b/include/linux/nd.h
index 255c38a83083..23276ea91690 100644
--- a/include/linux/nd.h
+++ b/include/linux/nd.h
@@ -50,6 +50,26 @@ struct nd_namespace_pmem {
u8 *uuid;
};
+/**
+ * struct nd_namespace_blk - namespace for dimm-bounded persistent memory
+ * @dev: namespace device creation by the nd region driver
+ * @alt_name: namespace name supplied in the dimm label
+ * @uuid: namespace name supplied in the dimm label
+ * @id: ida allocated id
+ * @lbasize: blk namespaces have a native sector size when btt not present
+ * @num_resources: number of dpa extents to claim
+ * @res: discontiguous dpa extents for given dimm
+ */
+struct nd_namespace_blk {
+ struct device dev;
+ char *alt_name;
+ u8 *uuid;
+ int id;
+ unsigned long lbasize;
+ int num_resources;
+ struct resource **res;
+};
+
static inline struct nd_namespace_io *to_nd_namespace_io(struct device *dev)
{
return container_of(dev, struct nd_namespace_io, dev);
@@ -62,6 +82,11 @@ static inline struct nd_namespace_pmem *to_nd_namespace_pmem(struct device *dev)
return container_of(nsio, struct nd_namespace_pmem, nsio);
}
+static inline struct nd_namespace_blk *to_nd_namespace_blk(struct device *dev)
+{
+ return container_of(dev, struct nd_namespace_blk, dev);
+}
+
#define MODULE_ALIAS_ND_DEVICE(type) \
MODULE_ALIAS("nd:t" __stringify(type) "*")
#define ND_DEVICE_MODALIAS_FMT "nd:t%d"
After 'uuid', 'size', and optionally 'alt_name' have been set to valid
values the labels on the dimms can be updated.
Write procedure is:
1/ Allocate and write new labels in the "next" index
2/ Free the old labels in the working copy
3/ Write the bitmap and the label space on the dimm
4/ Write the index to make the update valid
Label ranges directly mirror the dpa resource values for the given
label_id of the namespace.
Cc: Greg KH <[email protected]>
Cc: Neil Brown <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/dimm_devs.c | 49 ++++++
drivers/block/nd/label.c | 327 +++++++++++++++++++++++++++++++++++++
drivers/block/nd/label.h | 6 +
drivers/block/nd/namespace_devs.c | 82 ++++++++-
drivers/block/nd/nd.h | 3
5 files changed, 453 insertions(+), 14 deletions(-)
diff --git a/drivers/block/nd/dimm_devs.c b/drivers/block/nd/dimm_devs.c
index 4aa5654354ac..358b2a06d680 100644
--- a/drivers/block/nd/dimm_devs.c
+++ b/drivers/block/nd/dimm_devs.c
@@ -132,6 +132,55 @@ int nd_dimm_init_config_data(struct nd_dimm_drvdata *ndd)
return rc;
}
+int nd_dimm_set_config_data(struct nd_dimm_drvdata *ndd, size_t offset,
+ void *buf, size_t len)
+{
+ int rc = validate_dimm(ndd);
+ size_t max_cmd_size, buf_offset;
+ struct nd_cmd_set_config_hdr *cmd;
+ struct nd_bus *nd_bus = walk_to_nd_bus(ndd->dev);
+ struct nd_bus_descriptor *nd_desc = nd_bus->nd_desc;
+
+ if (rc)
+ return rc;
+
+ if (!ndd->data)
+ return -ENXIO;
+
+ if (offset + len > ndd->nsarea.config_size)
+ return -ENXIO;
+
+ max_cmd_size = min_t(u32, PAGE_SIZE, len);
+ max_cmd_size = min_t(u32, max_cmd_size, ndd->nsarea.max_xfer);
+ cmd = kzalloc(max_cmd_size + sizeof(*cmd) + sizeof(u32), GFP_KERNEL);
+ if (!cmd)
+ return -ENOMEM;
+
+ for (buf_offset = 0; len; len -= cmd->in_length,
+ buf_offset += cmd->in_length) {
+ size_t cmd_size;
+ u32 *status;
+
+ cmd->in_offset = offset + buf_offset;
+ cmd->in_length = min(max_cmd_size, len);
+ memcpy(cmd->in_buf, buf + buf_offset, cmd->in_length);
+
+ /* status is output in the last 4-bytes of the command buffer */
+ cmd_size = sizeof(*cmd) + cmd->in_length + sizeof(u32);
+ status = ((void *) cmd) + cmd_size - sizeof(u32);
+
+ rc = nd_desc->ndctl(nd_desc, to_nd_dimm(ndd->dev),
+ ND_CMD_SET_CONFIG_DATA, cmd, cmd_size);
+ if (rc || *status) {
+ rc = rc ? rc : -ENXIO;
+ break;
+ }
+ }
+ kfree(cmd);
+
+ return rc;
+}
+
static void nd_dimm_release(struct device *dev)
{
struct nd_dimm *nd_dimm = to_nd_dimm(dev);
diff --git a/drivers/block/nd/label.c b/drivers/block/nd/label.c
index b55fa2a6f872..78898b642191 100644
--- a/drivers/block/nd/label.c
+++ b/drivers/block/nd/label.c
@@ -12,6 +12,7 @@
*/
#include <linux/device.h>
#include <linux/ndctl.h>
+#include <linux/slab.h>
#include <linux/io.h>
#include <linux/nd.h>
#include "nd-private.h"
@@ -57,6 +58,11 @@ size_t sizeof_namespace_index(struct nd_dimm_drvdata *ndd)
return ndd->nsindex_size;
}
+static int nd_dimm_num_label_slots(struct nd_dimm_drvdata *ndd)
+{
+ return ndd->nsarea.config_size / 129;
+}
+
int nd_label_validate(struct nd_dimm_drvdata *ndd)
{
/*
@@ -202,23 +208,30 @@ static struct nd_namespace_label __iomem *nd_label_base(struct nd_dimm_drvdata *
return base + 2 * sizeof_namespace_index(ndd);
}
+static int to_slot(struct nd_dimm_drvdata *ndd,
+ struct nd_namespace_label __iomem *nd_label)
+{
+ return nd_label - nd_label_base(ndd);
+}
+
#define for_each_clear_bit_le(bit, addr, size) \
for ((bit) = find_next_zero_bit_le((addr), (size), 0); \
(bit) < (size); \
(bit) = find_next_zero_bit_le((addr), (size), (bit) + 1))
/**
- * preamble_current - common variable initialization for nd_label_* routines
+ * preamble_index - common variable initialization for nd_label_* routines
* @nd_dimm: dimm container for the relevant label set
+ * @idx: namespace_index index
* @nsindex: on return set to the currently active namespace index
* @free: on return set to the free label bitmap in the index
* @nslot: on return set to the number of slots in the label space
*/
-static bool preamble_current(struct nd_dimm_drvdata *ndd,
+static bool preamble_index(struct nd_dimm_drvdata *ndd, int idx,
struct nd_namespace_index **nsindex,
unsigned long **free, u32 *nslot)
{
- *nsindex = to_current_namespace_index(ndd);
+ *nsindex = to_namespace_index(ndd, idx);
if (*nsindex == NULL)
return false;
@@ -237,6 +250,22 @@ char *nd_label_gen_id(struct nd_label_id *label_id, u8 *uuid, u32 flags)
return label_id->id;
}
+static bool preamble_current(struct nd_dimm_drvdata *ndd,
+ struct nd_namespace_index **nsindex,
+ unsigned long **free, u32 *nslot)
+{
+ return preamble_index(ndd, ndd->ns_current, nsindex,
+ free, nslot);
+}
+
+static bool preamble_next(struct nd_dimm_drvdata *ndd,
+ struct nd_namespace_index **nsindex,
+ unsigned long **free, u32 *nslot)
+{
+ return preamble_index(ndd, ndd->ns_next, nsindex,
+ free, nslot);
+}
+
static bool slot_valid(struct nd_namespace_label __iomem *nd_label, u32 slot)
{
/* check that we are written where we expect to be written */
@@ -341,3 +370,295 @@ struct nd_namespace_label __iomem *nd_label_active(
return NULL;
}
+
+u32 nd_label_alloc_slot(struct nd_dimm_drvdata *ndd)
+{
+ struct nd_namespace_index __iomem *nsindex;
+ unsigned long *free;
+ u32 nslot, slot;
+
+ if (!preamble_next(ndd, &nsindex, &free, &nslot))
+ return UINT_MAX;
+
+ WARN_ON(!is_nd_bus_locked(ndd->dev));
+
+ slot = find_next_bit_le(free, nslot, 0);
+ if (slot == nslot)
+ return UINT_MAX;
+
+ clear_bit_le(slot, free);
+
+ return slot;
+}
+
+bool nd_label_free_slot(struct nd_dimm_drvdata *ndd, u32 slot)
+{
+ struct nd_namespace_index __iomem *nsindex;
+ unsigned long *free;
+ u32 nslot;
+
+ if (!preamble_next(ndd, &nsindex, &free, &nslot))
+ return false;
+
+ WARN_ON(!is_nd_bus_locked(ndd->dev));
+
+ if (slot < nslot)
+ return !test_and_set_bit_le(slot, free);
+ return false;
+}
+
+u32 nd_label_nfree(struct nd_dimm_drvdata *ndd)
+{
+ struct nd_namespace_index __iomem *nsindex;
+ unsigned long *free;
+ u32 nslot;
+
+ WARN_ON(!is_nd_bus_locked(ndd->dev));
+
+ if (!preamble_next(ndd, &nsindex, &free, &nslot))
+ return 0;
+
+ return bitmap_weight(free, nslot);
+}
+
+static int nd_label_write_index(struct nd_dimm_drvdata *ndd, int index, u32 seq,
+ unsigned long flags)
+{
+ struct nd_namespace_index *nsindex = to_namespace_index(ndd, index);
+ unsigned long offset;
+ u64 checksum;
+ u32 nslot;
+ int rc;
+
+ if (flags & ND_NSINDEX_INIT)
+ nslot = nd_dimm_num_label_slots(ndd);
+ else
+ nslot = readl(&nsindex->nslot);
+
+ memcpy_toio(nsindex->sig, NSINDEX_SIGNATURE, NSINDEX_SIG_LEN);
+ writel(0, &nsindex->flags);
+ writel(seq, &nsindex->seq);
+ offset = (unsigned long) nsindex
+ - (unsigned long) to_namespace_index(ndd, 0);
+ writeq(offset, &nsindex->myoff);
+ writeq(sizeof_namespace_index(ndd), &nsindex->mysize);
+ offset = (unsigned long) to_namespace_index(ndd,
+ nd_label_next_nsindex(index))
+ - (unsigned long) to_namespace_index(ndd, 0);
+ writeq(offset, &nsindex->otheroff);
+ offset = (unsigned long) nd_label_base(ndd)
+ - (unsigned long) to_namespace_index(ndd, 0);
+ writeq(offset, &nsindex->labeloff);
+ writel(nslot, &nsindex->nslot);
+ writew(1, &nsindex->major);
+ writew(1, &nsindex->minor);
+ writeq(0, &nsindex->checksum);
+ if (flags & ND_NSINDEX_INIT) {
+ unsigned long *free = (unsigned long __force *) nsindex->free;
+ u32 nfree = ALIGN(nslot, BITS_PER_LONG);
+ int last_bits, i;
+
+ memset_io(nsindex->free, 0xff, nfree / 8);
+ for (i = 0, last_bits = nfree - nslot; i < last_bits; i++)
+ clear_bit_le(nslot + i, free);
+ }
+ checksum = nd_fletcher64(nsindex, sizeof_namespace_index(ndd));
+ writeq(checksum, &nsindex->checksum);
+ rc = nd_dimm_set_config_data(ndd, readq(&nsindex->myoff),
+ nsindex, sizeof_namespace_index(ndd));
+ if (rc < 0)
+ return rc;
+
+ if (flags & ND_NSINDEX_INIT)
+ return 0;
+
+ /* copy the index we just wrote to the new 'next' */
+ WARN_ON(index != ndd->ns_next);
+ nd_label_copy(ndd, to_current_namespace_index(ndd), nsindex);
+ ndd->ns_current = nd_label_next_nsindex(ndd->ns_current);
+ ndd->ns_next = nd_label_next_nsindex(ndd->ns_next);
+ WARN_ON(ndd->ns_current == ndd->ns_next);
+
+ return 0;
+}
+
+static unsigned long nd_label_offset(struct nd_dimm_drvdata *ndd,
+ struct nd_namespace_label __iomem *nd_label)
+{
+ return (unsigned long) nd_label
+ - (unsigned long) to_namespace_index(ndd, 0);
+}
+
+static int __pmem_label_update(struct nd_region *nd_region,
+ struct nd_mapping *nd_mapping, struct nd_namespace_pmem *nspm,
+ int pos)
+{
+ u64 cookie = nd_region_interleave_set_cookie(nd_region), rawsize;
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ struct nd_namespace_label __iomem *victim_label;
+ struct nd_namespace_label __iomem *nd_label;
+ struct nd_namespace_index __iomem *nsindex;
+ unsigned long *free;
+ u32 nslot, slot;
+ size_t offset;
+ int rc;
+
+ if (!preamble_next(ndd, &nsindex, &free, &nslot))
+ return -ENXIO;
+
+ /* allocate and write the label to the staging (next) index */
+ slot = nd_label_alloc_slot(ndd);
+ if (slot == UINT_MAX)
+ return -ENXIO;
+ dev_dbg(ndd->dev, "%s: allocated: %d\n", __func__, slot);
+
+ nd_label = nd_label_base(ndd) + slot;
+ memset_io(nd_label, 0, sizeof(struct nd_namespace_label));
+ memcpy_toio(nd_label->uuid, nspm->uuid, NSLABEL_UUID_LEN);
+ if (nspm->alt_name)
+ memcpy_toio(nd_label->name, nspm->alt_name, NSLABEL_NAME_LEN);
+ writel(NSLABEL_FLAG_UPDATING, &nd_label->flags);
+ writew(nd_region->ndr_mappings, &nd_label->nlabel);
+ writew(pos, &nd_label->position);
+ writeq(cookie, &nd_label->isetcookie);
+ rawsize = div_u64(resource_size(&nspm->nsio.res),
+ nd_region->ndr_mappings);
+ writeq(rawsize, &nd_label->rawsize);
+ writeq(nd_mapping->start, &nd_label->dpa);
+ writel(slot, &nd_label->slot);
+
+ /* update label */
+ offset = nd_label_offset(ndd, nd_label);
+ rc = nd_dimm_set_config_data(ndd, offset, nd_label,
+ sizeof(struct nd_namespace_label));
+ if (rc < 0)
+ return rc;
+
+ /* Garbage collect the previous label */
+ victim_label = nd_get_label(nd_mapping->labels, 0);
+ if (victim_label) {
+ slot = to_slot(ndd, victim_label);
+ nd_label_free_slot(ndd, slot);
+ dev_dbg(ndd->dev, "%s: free: %d\n", __func__, slot);
+ }
+
+ /* update index */
+ rc = nd_label_write_index(ndd, ndd->ns_next,
+ nd_inc_seq(readl(&nsindex->seq)), 0);
+ if (rc < 0)
+ return rc;
+
+ nd_set_label(nd_mapping->labels, nd_label, 0);
+
+ return 0;
+}
+
+static int init_labels(struct nd_mapping *nd_mapping)
+{
+ int i;
+ struct nd_namespace_index __iomem *nsindex;
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+
+ if (!nd_mapping->labels)
+ nd_mapping->labels = kcalloc(2, sizeof(void *), GFP_KERNEL);
+
+ if (!nd_mapping->labels)
+ return -ENOMEM;
+
+ if (ndd->ns_current == -1 || ndd->ns_next == -1)
+ /* pass */;
+ else
+ return 0;
+
+ nsindex = to_namespace_index(ndd, 0);
+ memset_io(nsindex, 0, ndd->nsarea.config_size);
+ for (i = 0; i < 2; i++) {
+ int rc = nd_label_write_index(ndd, i, i*2, ND_NSINDEX_INIT);
+
+ if (rc)
+ return rc;
+ }
+ ndd->ns_next = 1;
+ ndd->ns_current = 0;
+
+ return 0;
+}
+
+static int del_labels(struct nd_mapping *nd_mapping, u8 *uuid)
+{
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ struct nd_namespace_label __iomem *nd_label;
+ struct nd_namespace_index __iomem *nsindex;
+ u8 label_uuid[NSLABEL_UUID_LEN];
+ int l, num_freed = 0;
+ unsigned long *free;
+ u32 nslot, slot;
+
+ if (!uuid)
+ return 0;
+
+ /* no index || no labels == nothing to delete */
+ if (!preamble_next(ndd, &nsindex, &free, &nslot)
+ || !nd_mapping->labels)
+ return 0;
+
+ for_each_label(l, nd_label, nd_mapping->labels) {
+ int j;
+
+ memcpy_fromio(label_uuid, nd_label->uuid, NSLABEL_UUID_LEN);
+ if (memcmp(label_uuid, uuid, NSLABEL_UUID_LEN) != 0)
+ continue;
+ slot = to_slot(ndd, nd_label);
+ nd_label_free_slot(ndd, slot);
+ dev_dbg(ndd->dev, "%s: free: %d\n", __func__, slot);
+ for (j = l; nd_get_label(nd_mapping->labels, j + 1); j++) {
+ struct nd_namespace_label __iomem *next_label;
+
+ next_label = nd_get_label(nd_mapping->labels, j + 1);
+ nd_set_label(nd_mapping->labels, next_label, j);
+ }
+ nd_set_label(nd_mapping->labels, NULL, j);
+ num_freed++;
+ }
+
+ if (num_freed > l) {
+ /*
+ * num_freed will only ever be > l when we delete the last
+ * label
+ */
+ kfree(nd_mapping->labels);
+ nd_mapping->labels = NULL;
+ dev_dbg(ndd->dev, "%s: no more labels\n", __func__);
+ }
+
+ return nd_label_write_index(ndd, ndd->ns_next,
+ nd_inc_seq(readl(&nsindex->seq)), 0);
+}
+
+int nd_pmem_namespace_label_update(struct nd_region *nd_region,
+ struct nd_namespace_pmem *nspm, resource_size_t size)
+{
+ int i;
+
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ int rc;
+
+ if (size == 0) {
+ rc = del_labels(nd_mapping, nspm->uuid);
+ if (rc)
+ return rc;
+ continue;
+ }
+
+ rc = init_labels(nd_mapping);
+ if (rc)
+ return rc;
+
+ rc = __pmem_label_update(nd_region, nd_mapping, nspm, i);
+ if (rc)
+ return rc;
+ }
+
+ return 0;
+}
diff --git a/drivers/block/nd/label.h b/drivers/block/nd/label.h
index 4436624f4146..e17958941e34 100644
--- a/drivers/block/nd/label.h
+++ b/drivers/block/nd/label.h
@@ -34,6 +34,7 @@ enum {
BTTINFO_MAJOR_VERSION = 1,
ND_LABEL_MIN_SIZE = 512 * 129, /* see sizeof_namespace_index() */
ND_LABEL_ID_SIZE = 50,
+ ND_NSINDEX_INIT = 0x1,
};
static const char NSINDEX_SIGNATURE[] = "NAMESPACE_INDEX\0";
@@ -129,4 +130,9 @@ size_t sizeof_namespace_index(struct nd_dimm_drvdata *ndd);
int nd_label_active_count(struct nd_dimm_drvdata *ndd);
struct nd_namespace_label __iomem *nd_label_active(
struct nd_dimm_drvdata *ndd, int n);
+u32 nd_label_nfree(struct nd_dimm_drvdata *ndd);
+struct nd_region;
+struct nd_namespace_pmem;
+int nd_pmem_namespace_label_update(struct nd_region *nd_region,
+ struct nd_namespace_pmem *nspm, resource_size_t size);
#endif /* __LABEL_H__ */
diff --git a/drivers/block/nd/namespace_devs.c b/drivers/block/nd/namespace_devs.c
index e229755509a2..c04f55fb7c7c 100644
--- a/drivers/block/nd/namespace_devs.c
+++ b/drivers/block/nd/namespace_devs.c
@@ -151,20 +151,52 @@ static resource_size_t nd_namespace_blk_size(struct nd_namespace_blk *nsblk)
return size;
}
+static int nd_namespace_label_update(struct nd_region *nd_region, struct device *dev)
+{
+ dev_WARN_ONCE(dev, dev->driver,
+ "namespace must be idle during label update\n");
+ if (dev->driver)
+ return 0;
+
+ /*
+ * Only allow label writes that will result in a valid namespace
+ * or deletion of an existing namespace.
+ */
+ if (is_namespace_pmem(dev)) {
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+ struct resource *res = &nspm->nsio.res;
+ resource_size_t size = resource_size(res);
+
+ if (size == 0 && nspm->uuid)
+ /* delete allocation */;
+ else if (!nspm->uuid)
+ return 0;
+
+ return nd_pmem_namespace_label_update(nd_region, nspm, size);
+ } else if (is_namespace_blk(dev)) {
+ /* TODO: implement blk labels */
+ return 0;
+ } else
+ return -ENXIO;
+}
+
static ssize_t alt_name_store(struct device *dev,
struct device_attribute *attr, const char *buf, size_t len)
{
+ struct nd_region *nd_region = to_nd_region(dev->parent);
ssize_t rc;
device_lock(dev);
nd_bus_lock(dev);
wait_nd_bus_probe_idle(dev);
rc = __alt_name_store(dev, buf, len);
+ if (rc >= 0)
+ rc = nd_namespace_label_update(nd_region, dev);
dev_dbg(dev, "%s: %s (%zd)\n", __func__, rc < 0 ? "fail" : "success", rc);
nd_bus_unlock(dev);
device_unlock(dev);
- return rc;
+ return rc < 0 ? rc : len;
}
static ssize_t alt_name_show(struct device *dev,
@@ -707,6 +739,7 @@ static ssize_t __size_store(struct device *dev, unsigned long long val)
static ssize_t size_store(struct device *dev,
struct device_attribute *attr, const char *buf, size_t len)
{
+ struct nd_region *nd_region = to_nd_region(dev->parent);
unsigned long long val;
u8 **uuid = NULL;
int rc;
@@ -719,6 +752,8 @@ static ssize_t size_store(struct device *dev,
nd_bus_lock(dev);
wait_nd_bus_probe_idle(dev);
rc = __size_store(dev, val);
+ if (rc >= 0)
+ rc = nd_namespace_label_update(nd_region, dev);
if (is_namespace_pmem(dev)) {
struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
@@ -742,7 +777,7 @@ static ssize_t size_store(struct device *dev,
nd_bus_unlock(dev);
device_unlock(dev);
- return rc ? rc : len;
+ return rc < 0 ? rc : len;
}
static ssize_t size_show(struct device *dev,
@@ -802,17 +837,34 @@ static int namespace_update_uuid(struct nd_region *nd_region,
u32 flags = is_namespace_blk(dev) ? NSLABEL_FLAG_LOCAL : 0;
struct nd_label_id old_label_id;
struct nd_label_id new_label_id;
- int i, rc;
+ int i;
- rc = nd_is_uuid_unique(dev, new_uuid) ? 0 : -EINVAL;
- if (rc) {
- kfree(new_uuid);
- return rc;
- }
+ if (!nd_is_uuid_unique(dev, new_uuid))
+ return -EINVAL;
if (*old_uuid == NULL)
goto out;
+ /*
+ * If we've already written a label with this uuid, then it's
+ * too late to rename because we can't reliably update the uuid
+ * without losing the old namespace. Userspace must delete this
+ * namespace to abandon the old uuid.
+ */
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+
+ /*
+ * This check by itself is sufficient because old_uuid
+ * would be NULL above if this uuid did not exist in the
+ * currently written set.
+ *
+ * FIXME: can we delete uuid with zero dpa allocated?
+ */
+ if (nd_mapping->labels)
+ return -EBUSY;
+ }
+
nd_label_gen_id(&old_label_id, *old_uuid, flags);
nd_label_gen_id(&new_label_id, new_uuid, flags);
for (i = 0; i < nd_region->ndr_mappings; i++) {
@@ -856,12 +908,16 @@ static ssize_t uuid_store(struct device *dev,
rc = nd_uuid_store(dev, &uuid, buf, len);
if (rc >= 0)
rc = namespace_update_uuid(nd_region, dev, uuid, ns_uuid);
+ if (rc >= 0)
+ rc = nd_namespace_label_update(nd_region, dev);
+ else
+ kfree(uuid);
dev_dbg(dev, "%s: result: %zd wrote: %s%s", __func__,
rc, buf, buf[len - 1] == '\n' ? "" : "\n");
nd_bus_unlock(dev);
device_unlock(dev);
- return rc ? rc : len;
+ return rc < 0 ? rc : len;
}
static DEVICE_ATTR_RW(uuid);
@@ -905,6 +961,7 @@ static ssize_t sector_size_store(struct device *dev,
struct device_attribute *attr, const char *buf, size_t len)
{
struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+ struct nd_region *nd_region = to_nd_region(dev->parent);
ssize_t rc;
if (!is_namespace_blk(dev))
@@ -914,8 +971,11 @@ static ssize_t sector_size_store(struct device *dev,
nd_bus_lock(dev);
rc = nd_sector_size_store(dev, buf, &nsblk->lbasize,
ns_lbasize_supported);
- dev_dbg(dev, "%s: result: %zd wrote: %s%s", __func__,
- rc, buf, buf[len - 1] == '\n' ? "" : "\n");
+ if (rc >= 0)
+ rc = nd_namespace_label_update(nd_region, dev);
+ dev_dbg(dev, "%s: result: %zd %s: %s%s", __func__,
+ rc, rc < 0 ? "tried" : "wrote", buf,
+ buf[len - 1] == '\n' ? "" : "\n");
nd_bus_unlock(dev);
device_unlock(dev);
diff --git a/drivers/block/nd/nd.h b/drivers/block/nd/nd.h
index 63d4c6c09275..8eaf0ae82db4 100644
--- a/drivers/block/nd/nd.h
+++ b/drivers/block/nd/nd.h
@@ -110,6 +110,7 @@ static inline unsigned nd_inc_seq(unsigned seq)
return next[seq & 3];
}
+
enum nd_async_mode {
ND_SYNC,
ND_ASYNC,
@@ -128,6 +129,8 @@ struct nd_dimm;
struct nd_dimm_drvdata *to_ndd(struct nd_mapping *nd_mapping);
int nd_dimm_init_nsarea(struct nd_dimm_drvdata *ndd);
int nd_dimm_init_config_data(struct nd_dimm_drvdata *ndd);
+int nd_dimm_set_config_data(struct nd_dimm_drvdata *ndd, size_t offset,
+ void *buf, size_t len);
struct nd_region *to_nd_region(struct device *dev);
int nd_region_to_namespace_type(struct nd_region *nd_region);
int nd_region_register_namespaces(struct nd_region *nd_region, int *err);
After 'uuid', 'size', 'sector_size', and optionally 'alt_name' have been
set to valid values the labels on the dimm can be updated. The
difference with the pmem case is that blk namespaces are limited to one
dimm and can cover discontiguous ranges in dpa space.
Also, after allocating label slots, it is useful for userspace to know
how many slots are left. Export this information in sysfs.
Cc: Greg KH <[email protected]>
Cc: Neil Brown <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/bus.c | 4
drivers/block/nd/dimm_devs.c | 25 +++
drivers/block/nd/label.c | 297 +++++++++++++++++++++++++++++++++++--
drivers/block/nd/label.h | 5 +
drivers/block/nd/namespace_devs.c | 57 +++++++
drivers/block/nd/nd-private.h | 1
6 files changed, 367 insertions(+), 22 deletions(-)
diff --git a/drivers/block/nd/bus.c b/drivers/block/nd/bus.c
index 819259e92468..6c272f245f4e 100644
--- a/drivers/block/nd/bus.c
+++ b/drivers/block/nd/bus.c
@@ -136,6 +136,10 @@ static void nd_async_device_unregister(void *d, async_cookie_t cookie)
{
struct device *dev = d;
+ /* flush bus operations before delete */
+ nd_bus_lock(dev);
+ nd_bus_unlock(dev);
+
device_unregister(dev);
put_device(dev);
}
diff --git a/drivers/block/nd/dimm_devs.c b/drivers/block/nd/dimm_devs.c
index 358b2a06d680..4b225c8b7d0a 100644
--- a/drivers/block/nd/dimm_devs.c
+++ b/drivers/block/nd/dimm_devs.c
@@ -19,6 +19,7 @@
#include <linux/fs.h>
#include <linux/mm.h>
#include "nd-private.h"
+#include "label.h"
#include "nd.h"
static DEFINE_IDA(dimm_ida);
@@ -262,9 +263,33 @@ static ssize_t state_show(struct device *dev, struct device_attribute *attr,
}
static DEVICE_ATTR_RO(state);
+static ssize_t available_slots_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_dimm_drvdata *ndd = dev_get_drvdata(dev);
+ ssize_t rc;
+ u32 nfree;
+
+ if (!ndd)
+ return -ENXIO;
+
+ nd_bus_lock(dev);
+ nfree = nd_label_nfree(ndd);
+ if (nfree - 1 > nfree) {
+ dev_WARN_ONCE(dev, 1, "we ate our last label?\n");
+ nfree = 0;
+ } else
+ nfree--;
+ rc = sprintf(buf, "%d\n", nfree);
+ nd_bus_unlock(dev);
+ return rc;
+}
+static DEVICE_ATTR_RO(available_slots);
+
static struct attribute *nd_dimm_attributes[] = {
&dev_attr_state.attr,
&dev_attr_commands.attr,
+ &dev_attr_available_slots.attr,
NULL,
};
diff --git a/drivers/block/nd/label.c b/drivers/block/nd/label.c
index 78898b642191..069c26d50ed1 100644
--- a/drivers/block/nd/label.c
+++ b/drivers/block/nd/label.c
@@ -58,7 +58,7 @@ size_t sizeof_namespace_index(struct nd_dimm_drvdata *ndd)
return ndd->nsindex_size;
}
-static int nd_dimm_num_label_slots(struct nd_dimm_drvdata *ndd)
+int nd_dimm_num_label_slots(struct nd_dimm_drvdata *ndd)
{
return ndd->nsarea.config_size / 129;
}
@@ -416,7 +416,7 @@ u32 nd_label_nfree(struct nd_dimm_drvdata *ndd)
WARN_ON(!is_nd_bus_locked(ndd->dev));
if (!preamble_next(ndd, &nsindex, &free, &nslot))
- return 0;
+ return nd_dimm_num_label_slots(ndd);
return bitmap_weight(free, nslot);
}
@@ -553,22 +553,270 @@ static int __pmem_label_update(struct nd_region *nd_region,
return 0;
}
-static int init_labels(struct nd_mapping *nd_mapping)
+static void del_label(struct nd_mapping *nd_mapping, int l)
+{
+ struct nd_namespace_label __iomem *next_label, __iomem *nd_label;
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ unsigned int slot;
+ int j;
+
+ nd_label = nd_get_label(nd_mapping->labels, l);
+ slot = to_slot(ndd, nd_label);
+ dev_vdbg(ndd->dev, "%s: clear: %d\n", __func__, slot);
+
+ for (j = l; (next_label = nd_get_label(nd_mapping->labels, j + 1)); j++)
+ nd_set_label(nd_mapping->labels, next_label, j);
+ nd_set_label(nd_mapping->labels, NULL, j);
+}
+
+static bool is_old_resource(struct resource *res, struct resource **list, int n)
{
int i;
+
+ if (res->flags & DPA_RESOURCE_ADJUSTED)
+ return false;
+ for (i = 0; i < n; i++)
+ if (res == list[i])
+ return true;
+ return false;
+}
+
+static struct resource *to_resource(struct nd_dimm_drvdata *ndd,
+ struct nd_namespace_label __iomem *nd_label)
+{
+ struct resource *res;
+
+ for_each_dpa_resource(ndd, res) {
+ if (res->start != readq(&nd_label->dpa))
+ continue;
+ if (resource_size(res) != readq(&nd_label->rawsize))
+ continue;
+ return res;
+ }
+
+ return NULL;
+}
+
+/*
+ * 1/ Account all the labels that can be freed after this update
+ * 2/ Allocate and write the label to the staging (next) index
+ * 3/ Record the resources in the namespace device
+ */
+static int __blk_label_update(struct nd_region *nd_region,
+ struct nd_mapping *nd_mapping, struct nd_namespace_blk *nsblk,
+ int num_labels)
+{
+ int i, l, alloc, victims, nfree, old_num_resources, nlabel, rc = -ENXIO;
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ struct nd_namespace_label __iomem *nd_label;
+ struct nd_namespace_index __iomem *nsindex;
+ unsigned long *free, *victim_map = NULL;
+ struct resource *res, **old_res_list;
+ struct nd_label_id label_id;
+ u8 uuid[NSLABEL_UUID_LEN];
+ u32 nslot, slot;
+
+ if (!preamble_next(ndd, &nsindex, &free, &nslot))
+ return -ENXIO;
+
+ old_res_list = nsblk->res;
+ nfree = nd_label_nfree(ndd);
+ old_num_resources = nsblk->num_resources;
+ nd_label_gen_id(&label_id, nsblk->uuid, NSLABEL_FLAG_LOCAL);
+
+ /*
+ * We need to loop over the old resources a few times, which seems a
+ * bit inefficient, but we need to know that we have the label
+ * space before we start mutating the tracking structures.
+ * Otherwise the recovery method of last resort for userspace is
+ * disable and re-enable the parent region.
+ */
+ alloc = 0;
+ for_each_dpa_resource(ndd, res) {
+ if (strcmp(res->name, label_id.id) != 0)
+ continue;
+ if (!is_old_resource(res, old_res_list, old_num_resources))
+ alloc++;
+ }
+
+ victims = 0;
+ if (old_num_resources) {
+ /* convert old local-label-map to dimm-slot victim-map */
+ victim_map = kcalloc(BITS_TO_LONGS(nslot), sizeof(long),
+ GFP_KERNEL);
+ if (!victim_map)
+ return -ENOMEM;
+
+ /* mark unused labels for garbage collection */
+ for_each_clear_bit_le(slot, free, nslot) {
+ nd_label = nd_label_base(ndd) + slot;
+ memcpy_fromio(uuid, nd_label->uuid, NSLABEL_UUID_LEN);
+ if (memcmp(uuid, nsblk->uuid, NSLABEL_UUID_LEN) != 0)
+ continue;
+ res = to_resource(ndd, nd_label);
+ if (res && is_old_resource(res, old_res_list,
+ old_num_resources))
+ continue;
+ slot = to_slot(ndd, nd_label);
+ set_bit(slot, victim_map);
+ victims++;
+ }
+ }
+
+ /* don't allow updates that consume the last label */
+ if (nfree - alloc < 0 || nfree - alloc + victims < 1) {
+ dev_info(&nsblk->dev, "insufficient label space\n");
+ kfree(victim_map);
+ return -ENOSPC;
+ }
+ /* from here on we need to abort on error */
+
+
+ /* assign all resources to the namespace before writing the labels */
+ nsblk->res = NULL;
+ nsblk->num_resources = 0;
+ for_each_dpa_resource(ndd, res) {
+ if (strcmp(res->name, label_id.id) != 0)
+ continue;
+ if (!nsblk_add_resource(nd_region, ndd, nsblk, res->start)) {
+ rc = -ENOMEM;
+ goto abort;
+ }
+ }
+
+ for (i = 0; i < nsblk->num_resources; i++) {
+ size_t offset;
+
+ res = nsblk->res[i];
+ if (is_old_resource(res, old_res_list, old_num_resources))
+ continue; /* carry-over */
+ slot = nd_label_alloc_slot(ndd);
+ if (slot == UINT_MAX)
+ goto abort;
+ dev_dbg(ndd->dev, "%s: allocated: %d\n", __func__, slot);
+
+ nd_label = nd_label_base(ndd) + slot;
+ memset_io(nd_label, 0, sizeof(struct nd_namespace_label));
+ memcpy_toio(nd_label->uuid, nsblk->uuid, NSLABEL_UUID_LEN);
+ if (nsblk->alt_name)
+ memcpy_toio(nd_label->name, nsblk->alt_name,
+ NSLABEL_NAME_LEN);
+ writel(NSLABEL_FLAG_LOCAL, &nd_label->flags);
+ writew(0, &nd_label->nlabel); /* N/A */
+ writew(0, &nd_label->position); /* N/A */
+ writeq(0, &nd_label->isetcookie); /* N/A */
+ writeq(res->start, &nd_label->dpa);
+ writeq(resource_size(res), &nd_label->rawsize);
+ writeq(nsblk->lbasize, &nd_label->lbasize);
+ writel(slot, &nd_label->slot);
+
+ /* update label */
+ offset = nd_label_offset(ndd, nd_label);
+ rc = nd_dimm_set_config_data(ndd, offset, nd_label,
+ sizeof(struct nd_namespace_label));
+ if (rc < 0)
+ goto abort;
+ }
+
+ /* free up now unused slots in the new index */
+ for_each_set_bit(slot, victim_map, victim_map ? nslot : 0) {
+ dev_dbg(ndd->dev, "%s: free: %d\n", __func__, slot);
+ nd_label_free_slot(ndd, slot);
+ }
+
+ /* update index */
+ rc = nd_label_write_index(ndd, ndd->ns_next,
+ nd_inc_seq(readl(&nsindex->seq)), 0);
+ if (rc)
+ goto abort;
+
+ /*
+ * Now that the on-dimm labels are up to date, fix up the tracking
+ * entries in nd_mapping->labels
+ */
+ nlabel = 0;
+ for_each_label(l, nd_label, nd_mapping->labels) {
+ nlabel++;
+ memcpy_fromio(uuid, nd_label->uuid, NSLABEL_UUID_LEN);
+ if (memcmp(uuid, nsblk->uuid, NSLABEL_UUID_LEN) != 0)
+ continue;
+ nlabel--;
+ del_label(nd_mapping, l);
+ l--; /* retry with the new label at this index */
+ }
+ if (nlabel + nsblk->num_resources > num_labels) {
+ /*
+ * Bug, we can't end up with more resources than
+ * available labels
+ */
+ WARN_ON_ONCE(1);
+ rc = -ENXIO;
+ goto out;
+ }
+
+ for_each_clear_bit_le(slot, free, nslot) {
+ nd_label = nd_label_base(ndd) + slot;
+ memcpy_fromio(uuid, nd_label->uuid, NSLABEL_UUID_LEN);
+ if (memcmp(uuid, nsblk->uuid, NSLABEL_UUID_LEN) != 0)
+ continue;
+ res = to_resource(ndd, nd_label);
+ res->flags &= ~DPA_RESOURCE_ADJUSTED;
+ dev_vdbg(&nsblk->dev, "assign label[%d] slot: %d\n", l, slot);
+ nd_set_label(nd_mapping->labels, nd_label, l++);
+ }
+ nd_set_label(nd_mapping->labels, NULL, l);
+
+ out:
+ kfree(old_res_list);
+ kfree(victim_map);
+ return rc;
+
+ abort:
+ /*
+ * 1/ repair the allocated label bitmap in the index
+ * 2/ restore the resource list
+ */
+ nd_label_copy(ndd, nsindex, to_current_namespace_index(ndd));
+ kfree(nsblk->res);
+ nsblk->res = old_res_list;
+ nsblk->num_resources = old_num_resources;
+ old_res_list = NULL;
+ goto out;
+}
+
+static int init_labels(struct nd_mapping *nd_mapping, int num_labels)
+{
+ int i, l, old_num_labels = 0;
struct nd_namespace_index __iomem *nsindex;
+ struct nd_namespace_label __iomem *nd_label;
struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ size_t size = (num_labels + 1) * sizeof(struct nd_namespace_label *);
- if (!nd_mapping->labels)
- nd_mapping->labels = kcalloc(2, sizeof(void *), GFP_KERNEL);
+ for_each_label(l, nd_label, nd_mapping->labels)
+ old_num_labels++;
+
+ /*
+ * We need to preserve all the old labels for the mapping so
+ * they can be garbage collected after writing the new labels.
+ */
+ if (num_labels > old_num_labels) {
+ struct nd_namespace_label **labels;
+ labels = krealloc(nd_mapping->labels, size, GFP_KERNEL);
+ if (!labels)
+ return -ENOMEM;
+ nd_mapping->labels = labels;
+ }
if (!nd_mapping->labels)
return -ENOMEM;
+ for (i = old_num_labels; i <= num_labels; i++)
+ nd_set_label(nd_mapping->labels, NULL, i);
+
if (ndd->ns_current == -1 || ndd->ns_next == -1)
/* pass */;
else
- return 0;
+ return max(num_labels, old_num_labels);
nsindex = to_namespace_index(ndd, 0);
memset_io(nsindex, 0, ndd->nsarea.config_size);
@@ -581,7 +829,7 @@ static int init_labels(struct nd_mapping *nd_mapping)
ndd->ns_next = 1;
ndd->ns_current = 0;
- return 0;
+ return max(num_labels, old_num_labels);
}
static int del_labels(struct nd_mapping *nd_mapping, u8 *uuid)
@@ -603,22 +851,15 @@ static int del_labels(struct nd_mapping *nd_mapping, u8 *uuid)
return 0;
for_each_label(l, nd_label, nd_mapping->labels) {
- int j;
-
memcpy_fromio(label_uuid, nd_label->uuid, NSLABEL_UUID_LEN);
if (memcmp(label_uuid, uuid, NSLABEL_UUID_LEN) != 0)
continue;
slot = to_slot(ndd, nd_label);
nd_label_free_slot(ndd, slot);
dev_dbg(ndd->dev, "%s: free: %d\n", __func__, slot);
- for (j = l; nd_get_label(nd_mapping->labels, j + 1); j++) {
- struct nd_namespace_label __iomem *next_label;
-
- next_label = nd_get_label(nd_mapping->labels, j + 1);
- nd_set_label(nd_mapping->labels, next_label, j);
- }
- nd_set_label(nd_mapping->labels, NULL, j);
+ del_label(nd_mapping, l);
num_freed++;
+ l--; /* retry with new label at this index */
}
if (num_freed > l) {
@@ -651,8 +892,8 @@ int nd_pmem_namespace_label_update(struct nd_region *nd_region,
continue;
}
- rc = init_labels(nd_mapping);
- if (rc)
+ rc = init_labels(nd_mapping, 1);
+ if (rc < 0)
return rc;
rc = __pmem_label_update(nd_region, nd_mapping, nspm, i);
@@ -662,3 +903,23 @@ int nd_pmem_namespace_label_update(struct nd_region *nd_region,
return 0;
}
+
+int nd_blk_namespace_label_update(struct nd_region *nd_region,
+ struct nd_namespace_blk *nsblk, resource_size_t size)
+{
+ struct nd_mapping *nd_mapping = &nd_region->mapping[0];
+ struct resource *res;
+ int count = 0;
+
+ if (size == 0)
+ return del_labels(nd_mapping, nsblk->uuid);
+
+ for_each_dpa_resource(to_ndd(nd_mapping), res)
+ count++;
+
+ count = init_labels(nd_mapping, count);
+ if (count < 0)
+ return count;
+
+ return __blk_label_update(nd_region, nd_mapping, nsblk, count);
+}
diff --git a/drivers/block/nd/label.h b/drivers/block/nd/label.h
index e17958941e34..a26cebc9f389 100644
--- a/drivers/block/nd/label.h
+++ b/drivers/block/nd/label.h
@@ -130,9 +130,14 @@ size_t sizeof_namespace_index(struct nd_dimm_drvdata *ndd);
int nd_label_active_count(struct nd_dimm_drvdata *ndd);
struct nd_namespace_label __iomem *nd_label_active(
struct nd_dimm_drvdata *ndd, int n);
+u32 nd_label_alloc_slot(struct nd_dimm_drvdata *ndd);
+bool nd_label_free_slot(struct nd_dimm_drvdata *ndd, u32 slot);
u32 nd_label_nfree(struct nd_dimm_drvdata *ndd);
struct nd_region;
struct nd_namespace_pmem;
+struct nd_namespace_blk;
int nd_pmem_namespace_label_update(struct nd_region *nd_region,
struct nd_namespace_pmem *nspm, resource_size_t size);
+int nd_blk_namespace_label_update(struct nd_region *nd_region,
+ struct nd_namespace_blk *nsblk, resource_size_t size);
#endif /* __LABEL_H__ */
diff --git a/drivers/block/nd/namespace_devs.c b/drivers/block/nd/namespace_devs.c
index c04f55fb7c7c..4dc499758636 100644
--- a/drivers/block/nd/namespace_devs.c
+++ b/drivers/block/nd/namespace_devs.c
@@ -164,8 +164,7 @@ static int nd_namespace_label_update(struct nd_region *nd_region, struct device
*/
if (is_namespace_pmem(dev)) {
struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
- struct resource *res = &nspm->nsio.res;
- resource_size_t size = resource_size(res);
+ resource_size_t size = resource_size(&nspm->nsio.res);
if (size == 0 && nspm->uuid)
/* delete allocation */;
@@ -174,8 +173,15 @@ static int nd_namespace_label_update(struct nd_region *nd_region, struct device
return nd_pmem_namespace_label_update(nd_region, nspm, size);
} else if (is_namespace_blk(dev)) {
- /* TODO: implement blk labels */
- return 0;
+ struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+ resource_size_t size = nd_namespace_blk_size(nsblk);
+
+ if (size == 0 && nsblk->uuid)
+ /* delete allocation */;
+ else if (!nsblk->uuid || !nsblk->lbasize)
+ return 0;
+
+ return nd_blk_namespace_label_update(nd_region, nsblk, size);
} else
return -ENXIO;
}
@@ -983,6 +989,48 @@ static ssize_t sector_size_store(struct device *dev,
}
static DEVICE_ATTR_RW(sector_size);
+static ssize_t dpa_extents_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_region *nd_region = to_nd_region(dev->parent);
+ struct nd_label_id label_id;
+ int count = 0, i;
+ u8 *uuid = NULL;
+ u32 flags = 0;
+
+ nd_bus_lock(dev);
+ if (is_namespace_pmem(dev)) {
+ struct nd_namespace_pmem *nspm = to_nd_namespace_pmem(dev);
+
+ uuid = nspm->uuid;
+ flags = 0;
+ } else if (is_namespace_blk(dev)) {
+ struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+
+ uuid = nsblk->uuid;
+ flags = NSLABEL_FLAG_LOCAL;
+ }
+
+ if (!uuid)
+ goto out;
+
+ nd_label_gen_id(&label_id, uuid, flags);
+ for (i = 0; i < nd_region->ndr_mappings; i++) {
+ struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ struct resource *res;
+
+ for_each_dpa_resource(ndd, res)
+ if (strcmp(res->name, label_id.id) == 0)
+ count++;
+ }
+ out:
+ nd_bus_unlock(dev);
+
+ return sprintf(buf, "%d\n", count);
+}
+static DEVICE_ATTR_RO(dpa_extents);
+
static struct attribute *nd_namespace_attributes[] = {
&dev_attr_type.attr,
&dev_attr_size.attr,
@@ -990,6 +1038,7 @@ static struct attribute *nd_namespace_attributes[] = {
&dev_attr_resource.attr,
&dev_attr_alt_name.attr,
&dev_attr_sector_size.attr,
+ &dev_attr_dpa_extents.attr,
NULL,
};
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
index 43dc6dd36b42..c7b29c6f21cb 100644
--- a/drivers/block/nd/nd-private.h
+++ b/drivers/block/nd/nd-private.h
@@ -82,4 +82,5 @@ struct nd_mapping;
struct resource *nsblk_add_resource(struct nd_region *nd_region,
struct nd_dimm_drvdata *ndd, struct nd_namespace_blk *nsblk,
resource_size_t start);
+int nd_dimm_num_label_slots(struct nd_dimm_drvdata *ndd);
#endif /* __ND_PRIVATE_H__ */
Block devices from an nd bus, in addition to accepting "struct bio"
based requests, also have the capability to perform byte-aligned
accesses. By default only the bio/block interface is used. However, if
another driver can make effective use of the byte-aligned capability it
can claim/disable the block interface and use the byte-aligned "nd_io"
interface.
The BTT driver is the intended first consumer of this mechanism to allow
layering atomic sector update guarantees on top of nd_io capable
nd-bus-block-devices.
Cc: Greg KH <[email protected]>
Cc: Neil Brown <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/Kconfig | 3
drivers/block/nd/Makefile | 1
drivers/block/nd/btt.h | 45 ++++
drivers/block/nd/btt_devs.c | 442 +++++++++++++++++++++++++++++++++++++++++
drivers/block/nd/bus.c | 128 ++++++++++++
drivers/block/nd/core.c | 79 +++++++
drivers/block/nd/nd-private.h | 28 +++
drivers/block/nd/nd.h | 94 +++++++++
drivers/block/nd/pmem.c | 29 +++
include/uapi/linux/ndctl.h | 2
10 files changed, 847 insertions(+), 4 deletions(-)
create mode 100644 drivers/block/nd/btt.h
create mode 100644 drivers/block/nd/btt_devs.c
diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
index c5eaf195734d..15896db4de37 100644
--- a/drivers/block/nd/Kconfig
+++ b/drivers/block/nd/Kconfig
@@ -95,4 +95,7 @@ config BLK_DEV_PMEM
Say Y if you want to use a NVDIMM described by NFIT
+config ND_BTT_DEVS
+ def_bool y
+
endif
diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
index d588f691163c..0c6d64b7a69d 100644
--- a/drivers/block/nd/Makefile
+++ b/drivers/block/nd/Makefile
@@ -32,3 +32,4 @@ libnd-y += region_devs.o
libnd-y += region.o
libnd-y += namespace_devs.o
libnd-y += label.o
+libnd-$(CONFIG_ND_BTT_DEVS) += btt_devs.o
diff --git a/drivers/block/nd/btt.h b/drivers/block/nd/btt.h
new file mode 100644
index 000000000000..e8f6d8e0ddd3
--- /dev/null
+++ b/drivers/block/nd/btt.h
@@ -0,0 +1,45 @@
+/*
+ * Block Translation Table library
+ * Copyright (c) 2014-2015, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+#ifndef _LINUX_BTT_H
+#define _LINUX_BTT_H
+
+#include <linux/types.h>
+
+#define BTT_SIG_LEN 16
+#define BTT_SIG "BTT_ARENA_INFO\0"
+
+struct btt_sb {
+ u8 signature[BTT_SIG_LEN];
+ u8 uuid[16];
+ u8 parent_uuid[16];
+ __le32 flags;
+ __le16 version_major;
+ __le16 version_minor;
+ __le32 external_lbasize;
+ __le32 external_nlba;
+ __le32 internal_lbasize;
+ __le32 internal_nlba;
+ __le32 nfree;
+ __le32 infosize;
+ __le64 nextoff;
+ __le64 dataoff;
+ __le64 mapoff;
+ __le64 logoff;
+ __le64 info2off;
+ u8 padding[3968];
+ __le64 checksum;
+};
+
+#endif
diff --git a/drivers/block/nd/btt_devs.c b/drivers/block/nd/btt_devs.c
new file mode 100644
index 000000000000..e6f0b8b999d8
--- /dev/null
+++ b/drivers/block/nd/btt_devs.c
@@ -0,0 +1,442 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+#include <linux/device.h>
+#include <linux/genhd.h>
+#include <linux/sizes.h>
+#include <linux/slab.h>
+#include <linux/fs.h>
+#include <linux/mm.h>
+#include "nd-private.h"
+#include "btt.h"
+#include "nd.h"
+
+static DEFINE_IDA(btt_ida);
+
+static void nd_btt_release(struct device *dev)
+{
+ struct nd_btt *nd_btt = to_nd_btt(dev);
+
+ dev_dbg(dev, "%s\n", __func__);
+ WARN_ON(nd_btt->backing_dev);
+ ndio_del_claim(nd_btt->ndio_claim);
+ ida_simple_remove(&btt_ida, nd_btt->id);
+ kfree(nd_btt->uuid);
+ kfree(nd_btt);
+}
+
+static struct device_type nd_btt_device_type = {
+ .name = "nd_btt",
+ .release = nd_btt_release,
+};
+
+bool is_nd_btt(struct device *dev)
+{
+ return dev->type == &nd_btt_device_type;
+}
+
+struct nd_btt *to_nd_btt(struct device *dev)
+{
+ struct nd_btt *nd_btt = container_of(dev, struct nd_btt, dev);
+
+ WARN_ON(!is_nd_btt(dev));
+ return nd_btt;
+}
+EXPORT_SYMBOL(to_nd_btt);
+
+static const unsigned long btt_lbasize_supported[] = { 512, 4096, 0 };
+
+static ssize_t sector_size_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_btt *nd_btt = to_nd_btt(dev);
+
+ return nd_sector_size_show(nd_btt->lbasize, btt_lbasize_supported, buf);
+}
+
+static ssize_t sector_size_store(struct device *dev,
+ struct device_attribute *attr, const char *buf, size_t len)
+{
+ struct nd_btt *nd_btt = to_nd_btt(dev);
+ ssize_t rc;
+
+ device_lock(dev);
+ nd_bus_lock(dev);
+ rc = nd_sector_size_store(dev, buf, &nd_btt->lbasize,
+ btt_lbasize_supported);
+ dev_dbg(dev, "%s: result: %zd wrote: %s%s", __func__,
+ rc, buf, buf[len - 1] == '\n' ? "" : "\n");
+ nd_bus_unlock(dev);
+ device_unlock(dev);
+
+ return rc ? rc : len;
+}
+static DEVICE_ATTR_RW(sector_size);
+
+static ssize_t uuid_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_btt *nd_btt = to_nd_btt(dev);
+
+ if (nd_btt->uuid)
+ return sprintf(buf, "%pUb\n", nd_btt->uuid);
+ return sprintf(buf, "\n");
+}
+
+static ssize_t uuid_store(struct device *dev,
+ struct device_attribute *attr, const char *buf, size_t len)
+{
+ struct nd_btt *nd_btt = to_nd_btt(dev);
+ ssize_t rc;
+
+ device_lock(dev);
+ rc = nd_uuid_store(dev, &nd_btt->uuid, buf, len);
+ dev_dbg(dev, "%s: result: %zd wrote: %s%s", __func__,
+ rc, buf, buf[len - 1] == '\n' ? "" : "\n");
+ device_unlock(dev);
+
+ return rc ? rc : len;
+}
+static DEVICE_ATTR_RW(uuid);
+
+static ssize_t backing_dev_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_btt *nd_btt = to_nd_btt(dev);
+ char name[BDEVNAME_SIZE];
+
+ if (nd_btt->backing_dev)
+ return sprintf(buf, "/dev/%s\n",
+ bdevname(nd_btt->backing_dev, name));
+ else
+ return sprintf(buf, "\n");
+}
+
+static const fmode_t nd_btt_devs_mode = FMODE_READ | FMODE_WRITE | FMODE_EXCL;
+
+static void nd_btt_ndio_notify_remove(struct nd_io_claim *ndio_claim)
+{
+ char bdev_name[BDEVNAME_SIZE];
+ struct nd_btt *nd_btt;
+
+ if (!ndio_claim || !ndio_claim->holder)
+ return;
+
+ nd_btt = to_nd_btt(ndio_claim->holder);
+ WARN_ON_ONCE(!is_nd_bus_locked(&nd_btt->dev));
+ dev_dbg(&nd_btt->dev, "%pf: %s: release /dev/%s\n",
+ __builtin_return_address(0), __func__,
+ bdevname(nd_btt->backing_dev, bdev_name));
+ blkdev_put(nd_btt->backing_dev, nd_btt_devs_mode);
+ nd_btt->backing_dev = NULL;
+
+ /*
+ * Once we've had our backing device removed we need to be fully
+ * reconfigured. The bus will have already created a new seed
+ * for this purpose, so now is a good time to clean up this
+ * stale nd_btt instance.
+ */
+ if (nd_btt->dev.driver)
+ nd_device_unregister(&nd_btt->dev, ND_ASYNC);
+ else {
+ ndio_del_claim(ndio_claim);
+ nd_btt->ndio_claim = NULL;
+ }
+}
+
+static ssize_t __backing_dev_store(struct device *dev,
+ struct device_attribute *attr, const char *buf, size_t len)
+{
+ struct nd_bus *nd_bus = walk_to_nd_bus(dev);
+ struct nd_btt *nd_btt = to_nd_btt(dev);
+ char bdev_name[BDEVNAME_SIZE];
+ struct block_device *bdev;
+ struct nd_io *ndio;
+ char *path;
+
+ if (dev->driver) {
+ dev_dbg(dev, "%s: -EBUSY\n", __func__);
+ return -EBUSY;
+ }
+
+ path = kstrndup(buf, len, GFP_KERNEL);
+ if (!path)
+ return -ENOMEM;
+
+ /* detach the backing device */
+ if (strcmp(strim(path), "") == 0) {
+ if (!nd_btt->backing_dev)
+ goto out;
+ nd_btt_ndio_notify_remove(nd_btt->ndio_claim);
+ goto out;
+ } else if (nd_btt->backing_dev) {
+ dev_dbg(dev, "backing_dev already set\n");
+ len = -EBUSY;
+ goto out;
+ }
+
+ bdev = blkdev_get_by_path(strim(path), nd_btt_devs_mode, nd_btt);
+ if (IS_ERR(bdev)) {
+ dev_dbg(dev, "open '%s' failed: %ld\n", strim(path),
+ PTR_ERR(bdev));
+ len = PTR_ERR(bdev);
+ goto out;
+ }
+
+ if (get_capacity(bdev->bd_disk) < SZ_16M / 512) {
+ blkdev_put(bdev, nd_btt_devs_mode);
+ len = -ENXIO;
+ goto out;
+ }
+
+ ndio = ndio_lookup(nd_bus, bdevname(bdev->bd_contains, bdev_name));
+ if (!ndio) {
+ dev_dbg(dev, "%s does not have an ndio interface\n",
+ strim(path));
+ blkdev_put(bdev, nd_btt_devs_mode);
+ len = -ENXIO;
+ goto out;
+ }
+
+ nd_btt->ndio_claim = ndio_add_claim(ndio, &nd_btt->dev,
+ nd_btt_ndio_notify_remove);
+ if (!nd_btt->ndio_claim) {
+ blkdev_put(bdev, nd_btt_devs_mode);
+ len = -ENOMEM;
+ goto out;
+ }
+
+ WARN_ON_ONCE(!is_nd_bus_locked(&nd_btt->dev));
+ nd_btt->backing_dev = bdev;
+
+ out:
+ kfree(path);
+ return len;
+}
+
+static ssize_t backing_dev_store(struct device *dev,
+ struct device_attribute *attr, const char *buf, size_t len)
+{
+ ssize_t rc;
+
+ nd_bus_lock(dev);
+ device_lock(dev);
+ rc = __backing_dev_store(dev, attr, buf, len);
+ dev_dbg(dev, "%s: result: %zd wrote: %s%s", __func__,
+ rc, buf, buf[len - 1] == '\n' ? "" : "\n");
+ device_unlock(dev);
+ nd_bus_unlock(dev);
+
+ return rc;
+}
+static DEVICE_ATTR_RW(backing_dev);
+
+static bool is_nd_btt_idle(struct device *dev)
+{
+ struct nd_bus *nd_bus = walk_to_nd_bus(dev);
+ struct nd_btt *nd_btt = to_nd_btt(dev);
+
+ if (nd_bus->nd_btt == nd_btt || dev->driver || nd_btt->backing_dev)
+ return false;
+ return true;
+}
+
+static ssize_t delete_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ /* return 1 if can be deleted */
+ return sprintf(buf, "%d\n", is_nd_btt_idle(dev));
+}
+
+static ssize_t delete_store(struct device *dev,
+ struct device_attribute *attr, const char *buf, size_t len)
+{
+ unsigned long val;
+
+ /* write 1 to delete */
+ if (kstrtoul(buf, 0, &val) != 0 || val != 1)
+ return -EINVAL;
+
+ /* prevent deletion while this btt is active, or is the current seed */
+ if (!is_nd_btt_idle(dev))
+ return -EBUSY;
+
+ /*
+ * userspace raced itself if device goes active here and it gets
+ * to keep the pieces
+ */
+ nd_device_unregister(dev, ND_ASYNC);
+
+ return len;
+}
+static DEVICE_ATTR_RW(delete);
+
+static struct attribute *nd_btt_attributes[] = {
+ &dev_attr_sector_size.attr,
+ &dev_attr_backing_dev.attr,
+ &dev_attr_delete.attr,
+ &dev_attr_uuid.attr,
+ NULL,
+};
+
+static struct attribute_group nd_btt_attribute_group = {
+ .attrs = nd_btt_attributes,
+};
+
+static const struct attribute_group *nd_btt_attribute_groups[] = {
+ &nd_btt_attribute_group,
+ &nd_device_attribute_group,
+ NULL,
+};
+
+static struct nd_btt *__nd_btt_create(struct nd_bus *nd_bus,
+ unsigned long lbasize, u8 *uuid)
+{
+ struct nd_btt *nd_btt = kzalloc(sizeof(*nd_btt), GFP_KERNEL);
+ struct device *dev;
+
+ if (!nd_btt)
+ return NULL;
+ nd_btt->id = ida_simple_get(&btt_ida, 0, 0, GFP_KERNEL);
+ if (nd_btt->id < 0) {
+ kfree(nd_btt);
+ return NULL;
+ }
+
+ nd_btt->lbasize = lbasize;
+ if (uuid)
+ uuid = kmemdup(uuid, 16, GFP_KERNEL);
+ nd_btt->uuid = uuid;
+ dev = &nd_btt->dev;
+ dev_set_name(dev, "btt%d", nd_btt->id);
+ dev->parent = &nd_bus->dev;
+ dev->type = &nd_btt_device_type;
+ dev->groups = nd_btt_attribute_groups;
+ return nd_btt;
+}
+
+struct nd_btt *nd_btt_create(struct nd_bus *nd_bus)
+{
+ struct nd_btt *nd_btt = __nd_btt_create(nd_bus, 0, NULL);
+
+ if (!nd_btt)
+ return NULL;
+ nd_device_register(&nd_btt->dev);
+ return nd_btt;
+}
+
+/*
+ * nd_btt_sb_checksum: compute checksum for btt info block
+ *
+ * Returns a fletcher64 checksum of everything in the given info block
+ * except the last field (since that's where the checksum lives).
+ */
+u64 nd_btt_sb_checksum(struct btt_sb *btt_sb)
+{
+ u64 sum, sum_save;
+
+ sum_save = btt_sb->checksum;
+ btt_sb->checksum = 0;
+ sum = nd_fletcher64(btt_sb, sizeof(*btt_sb));
+ btt_sb->checksum = sum_save;
+ return sum;
+}
+EXPORT_SYMBOL(nd_btt_sb_checksum);
+
+static int nd_btt_autodetect(struct nd_bus *nd_bus, struct nd_io *ndio,
+ struct block_device *bdev)
+{
+ char name[BDEVNAME_SIZE];
+ struct nd_btt *nd_btt;
+ struct btt_sb *btt_sb;
+ u64 offset, checksum;
+ u32 lbasize;
+ u8 *uuid;
+ int rc;
+
+ btt_sb = kzalloc(sizeof(*btt_sb), GFP_KERNEL);
+ if (!btt_sb)
+ return -ENODEV;
+
+ offset = nd_partition_offset(bdev);
+ rc = ndio->rw_bytes(ndio, btt_sb, offset + SZ_4K, sizeof(*btt_sb), READ);
+ if (rc)
+ goto out_free_sb;
+
+ if (get_capacity(bdev->bd_disk) < SZ_16M / 512)
+ goto out_free_sb;
+
+ if (memcmp(btt_sb->signature, BTT_SIG, BTT_SIG_LEN) != 0)
+ goto out_free_sb;
+
+ checksum = le64_to_cpu(btt_sb->checksum);
+ btt_sb->checksum = 0;
+ if (checksum != nd_btt_sb_checksum(btt_sb))
+ goto out_free_sb;
+ btt_sb->checksum = cpu_to_le64(checksum);
+
+ uuid = kmemdup(btt_sb->uuid, 16, GFP_KERNEL);
+ if (!uuid)
+ goto out_free_sb;
+
+ lbasize = le32_to_cpu(btt_sb->external_lbasize);
+ nd_btt = __nd_btt_create(nd_bus, lbasize, uuid);
+ if (!nd_btt)
+ goto out_free_uuid;
+
+ device_initialize(&nd_btt->dev);
+ nd_btt->ndio_claim = ndio_add_claim(ndio, &nd_btt->dev,
+ nd_btt_ndio_notify_remove);
+ if (!nd_btt->ndio_claim)
+ goto out_free_btt;
+
+ nd_btt->backing_dev = bdev;
+ dev_dbg(&nd_btt->dev, "%s: activate %s\n", __func__,
+ bdevname(bdev, name));
+ __nd_device_register(&nd_btt->dev);
+ kfree(btt_sb);
+ return 0;
+
+ out_free_btt:
+ kfree(nd_btt);
+ out_free_uuid:
+ kfree(uuid);
+ out_free_sb:
+ kfree(btt_sb);
+
+ return -ENODEV;
+}
+
+void nd_btt_notify_ndio(struct nd_bus *nd_bus, struct nd_io *ndio)
+{
+ struct disk_part_iter piter;
+ struct hd_struct *part;
+
+ disk_part_iter_init(&piter, ndio->disk, DISK_PITER_INCL_PART0);
+ while ((part = disk_part_iter_next(&piter))) {
+ struct block_device *bdev;
+ int rc;
+
+ bdev = bdget_disk(ndio->disk, part->partno);
+ if (!bdev)
+ continue;
+ if (blkdev_get(bdev, nd_btt_devs_mode, nd_bus) != 0)
+ continue;
+ rc = nd_btt_autodetect(nd_bus, ndio, bdev);
+ if (rc)
+ blkdev_put(bdev, nd_btt_devs_mode);
+ /* no need to scan further in the case of whole disk btt */
+ if (rc == 0 && part->partno == 0)
+ break;
+ }
+ disk_part_iter_exit(&piter);
+}
diff --git a/drivers/block/nd/bus.c b/drivers/block/nd/bus.c
index 6c272f245f4e..27785f3fc4ca 100644
--- a/drivers/block/nd/bus.c
+++ b/drivers/block/nd/bus.c
@@ -16,6 +16,7 @@
#include <linux/module.h>
#include <linux/fcntl.h>
#include <linux/async.h>
+#include <linux/genhd.h>
#include <linux/ndctl.h>
#include <linux/sched.h>
#include <linux/slab.h>
@@ -40,6 +41,8 @@ static int to_nd_device_type(struct device *dev)
return ND_DEVICE_REGION_BLK;
else if (is_nd_pmem(dev->parent) || is_nd_blk(dev->parent))
return nd_region_to_namespace_type(to_nd_region(dev->parent));
+ else if (is_nd_btt(dev))
+ return ND_DEVICE_BTT;
return 0;
}
@@ -84,6 +87,21 @@ static int nd_bus_probe(struct device *dev)
dev_dbg(&nd_bus->dev, "%s.probe(%s) = %d\n", dev->driver->name,
dev_name(dev), rc);
+
+ /* check if our btt-seed has sprouted, and plant another */
+ if (rc == 0 && is_nd_btt(dev) && dev == &nd_bus->nd_btt->dev) {
+ const char *sep = "", *name = "", *status = "failed";
+
+ nd_bus->nd_btt = nd_btt_create(nd_bus);
+ if (nd_bus->nd_btt) {
+ status = "succeeded";
+ sep = ": ";
+ name = dev_name(&nd_bus->nd_btt->dev);
+ }
+ dev_dbg(&nd_bus->dev, "btt seed creation %s%s%s\n",
+ status, sep, name);
+ }
+
if (rc != 0)
module_put(provider);
return rc;
@@ -144,14 +162,19 @@ static void nd_async_device_unregister(void *d, async_cookie_t cookie)
put_device(dev);
}
-void nd_device_register(struct device *dev)
+void __nd_device_register(struct device *dev)
{
dev->bus = &nd_bus_type;
- device_initialize(dev);
get_device(dev);
async_schedule_domain(nd_async_device_register, dev,
&nd_async_domain);
}
+
+void nd_device_register(struct device *dev)
+{
+ device_initialize(dev);
+ __nd_device_register(dev);
+}
EXPORT_SYMBOL(nd_device_register);
void nd_device_unregister(struct device *dev, enum nd_async_mode mode)
@@ -200,6 +223,107 @@ int __nd_driver_register(struct nd_device_driver *nd_drv, struct module *owner,
}
EXPORT_SYMBOL(__nd_driver_register);
+/**
+ * nd_register_ndio() - register byte-aligned access capability for an nd-bdev
+ * @disk: child gendisk of the ndio namepace device
+ * @ndio: initialized ndio instance to register
+ *
+ * LOCKING: hold nd_bus_lock() over the creation of ndio->disk and the
+ * subsequent nd_region_ndio event
+ */
+int nd_register_ndio(struct nd_io *ndio)
+{
+ struct nd_bus *nd_bus;
+ struct device *dev;
+
+ if (!ndio || !ndio->dev || !ndio->disk || !list_empty(&ndio->list)
+ || !ndio->rw_bytes || !list_empty(&ndio->claims)) {
+ pr_debug("%s bad parameters from %pf\n", __func__,
+ __builtin_return_address(0));
+ return -EINVAL;
+ }
+
+ dev = ndio->dev;
+ nd_bus = walk_to_nd_bus(dev);
+ if (!nd_bus)
+ return -EINVAL;
+
+ WARN_ON_ONCE(!is_nd_bus_locked(&nd_bus->dev));
+ list_add(&ndio->list, &nd_bus->ndios);
+
+ /* TODO: generic infrastructure for 3rd party ndio claimers */
+ nd_btt_notify_ndio(nd_bus, ndio);
+
+ return 0;
+}
+EXPORT_SYMBOL(nd_register_ndio);
+
+/**
+ * __nd_unregister_ndio() - try to remove an ndio interface
+ * @ndio: interface to remove
+ */
+static int __nd_unregister_ndio(struct nd_io *ndio)
+{
+ struct nd_io_claim *ndio_claim, *_n;
+ struct nd_bus *nd_bus;
+ LIST_HEAD(claims);
+
+ nd_bus = walk_to_nd_bus(ndio->dev);
+ if (!nd_bus || list_empty(&ndio->list))
+ return -ENXIO;
+
+ spin_lock(&ndio->lock);
+ list_splice_init(&ndio->claims, &claims);
+ spin_unlock(&ndio->lock);
+
+ list_for_each_entry_safe(ndio_claim, _n, &claims, list)
+ ndio_claim->notify_remove(ndio_claim);
+
+ list_del_init(&ndio->list);
+
+ return 0;
+}
+
+int nd_unregister_ndio(struct nd_io *ndio)
+{
+ struct device *dev = ndio->dev;
+ int rc;
+
+ nd_bus_lock(dev);
+ rc = __nd_unregister_ndio(ndio);
+ nd_bus_unlock(dev);
+
+ /*
+ * Flush in case ->notify_remove() kicked off asynchronous device
+ * unregistration
+ */
+ nd_synchronize();
+
+ return rc;
+}
+EXPORT_SYMBOL(nd_unregister_ndio);
+
+static struct nd_io *__ndio_lookup(struct nd_bus *nd_bus, const char *diskname)
+{
+ struct nd_io *ndio;
+
+ list_for_each_entry(ndio, &nd_bus->ndios, list)
+ if (strcmp(diskname, ndio->disk->disk_name) == 0)
+ return ndio;
+
+ return NULL;
+}
+
+struct nd_io *ndio_lookup(struct nd_bus *nd_bus, const char *diskname)
+{
+ struct nd_io *ndio;
+
+ WARN_ON_ONCE(!is_nd_bus_locked(&nd_bus->dev));
+ ndio = __ndio_lookup(nd_bus, diskname);
+
+ return ndio;
+}
+
static ssize_t modalias_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
diff --git a/drivers/block/nd/core.c b/drivers/block/nd/core.c
index 3ec38289be58..dca48f9c1580 100644
--- a/drivers/block/nd/core.c
+++ b/drivers/block/nd/core.c
@@ -55,6 +55,62 @@ bool is_nd_bus_locked(struct device *dev)
}
EXPORT_SYMBOL(is_nd_bus_locked);
+void nd_init_ndio(struct nd_io *ndio, nd_rw_bytes_fn rw_bytes,
+ struct device *dev, struct gendisk *disk, unsigned long align)
+{
+ memset(ndio, 0, sizeof(*ndio));
+ INIT_LIST_HEAD(&ndio->claims);
+ INIT_LIST_HEAD(&ndio->list);
+ spin_lock_init(&ndio->lock);
+ ndio->dev = dev;
+ ndio->disk = disk;
+ ndio->align = align;
+ ndio->rw_bytes = rw_bytes;
+}
+EXPORT_SYMBOL(nd_init_ndio);
+
+void ndio_del_claim(struct nd_io_claim *ndio_claim)
+{
+ struct nd_io *ndio;
+ struct device *holder;
+
+ if (!ndio_claim)
+ return;
+ ndio = ndio_claim->parent;
+ holder = ndio_claim->holder;
+
+ dev_dbg(holder, "%s: drop %s\n", __func__, dev_name(ndio->dev));
+ spin_lock(&ndio->lock);
+ list_del(&ndio_claim->list);
+ spin_unlock(&ndio->lock);
+ put_device(ndio->dev);
+ kfree(ndio_claim);
+ put_device(holder);
+}
+
+struct nd_io_claim *ndio_add_claim(struct nd_io *ndio, struct device *holder,
+ ndio_notify_remove_fn notify_remove)
+{
+ struct nd_io_claim *ndio_claim = kzalloc(sizeof(*ndio_claim), GFP_KERNEL);
+
+ if (!ndio_claim)
+ return NULL;
+
+ INIT_LIST_HEAD(&ndio_claim->list);
+ ndio_claim->parent = ndio;
+ get_device(ndio->dev);
+
+ spin_lock(&ndio->lock);
+ list_add(&ndio_claim->list, &ndio->claims);
+ spin_unlock(&ndio->lock);
+
+ ndio_claim->holder = holder;
+ ndio_claim->notify_remove = notify_remove;
+ get_device(holder);
+
+ return ndio_claim;
+}
+
u64 nd_fletcher64(void __iomem *addr, size_t len)
{
u32 lo32 = 0;
@@ -74,6 +130,8 @@ static void nd_bus_release(struct device *dev)
{
struct nd_bus *nd_bus = container_of(dev, struct nd_bus, dev);
+ WARN_ON(!list_empty(&nd_bus->ndios));
+
ida_simple_remove(&nd_ida, nd_bus->id);
kfree(nd_bus);
}
@@ -270,10 +328,28 @@ static ssize_t wait_probe_show(struct device *dev,
}
static DEVICE_ATTR_RO(wait_probe);
+static ssize_t btt_seed_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct nd_bus *nd_bus = to_nd_bus(dev);
+ ssize_t rc;
+
+ nd_bus_lock(dev);
+ if (nd_bus->nd_btt)
+ rc = sprintf(buf, "%s\n", dev_name(&nd_bus->nd_btt->dev));
+ else
+ rc = sprintf(buf, "\n");
+ nd_bus_unlock(dev);
+
+ return rc;
+}
+static DEVICE_ATTR_RO(btt_seed);
+
static struct attribute *nd_bus_attributes[] = {
&dev_attr_commands.attr,
&dev_attr_wait_probe.attr,
&dev_attr_provider.attr,
+ &dev_attr_btt_seed.attr,
NULL,
};
@@ -290,6 +366,7 @@ struct nd_bus *__nd_bus_register(struct device *parent,
if (!nd_bus)
return NULL;
+ INIT_LIST_HEAD(&nd_bus->ndios);
INIT_LIST_HEAD(&nd_bus->list);
init_waitqueue_head(&nd_bus->probe_wait);
nd_bus->id = ida_simple_get(&nd_ida, 0, 0, GFP_KERNEL);
@@ -318,6 +395,8 @@ struct nd_bus *__nd_bus_register(struct device *parent,
list_add_tail(&nd_bus->list, &nd_bus_list);
mutex_unlock(&nd_bus_list_mutex);
+ nd_bus->nd_btt = nd_btt_create(nd_bus);
+
return nd_bus;
err:
put_device(&nd_bus->dev);
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
index c7b29c6f21cb..4ab4765dc9ee 100644
--- a/drivers/block/nd/nd-private.h
+++ b/drivers/block/nd/nd-private.h
@@ -22,14 +22,21 @@ extern struct list_head nd_bus_list;
extern struct mutex nd_bus_list_mutex;
extern int nd_dimm_major;
+struct block_device;
+struct nd_io_claim;
+struct nd_btt;
+struct nd_io;
+
struct nd_bus {
struct nd_bus_descriptor *nd_desc;
wait_queue_head_t probe_wait;
struct module *module;
+ struct list_head ndios;
struct list_head list;
struct device dev;
int id, probe_active;
struct mutex reconfig_mutex;
+ struct nd_btt *nd_btt;
};
struct nd_dimm {
@@ -41,9 +48,29 @@ struct nd_dimm {
int id;
};
+struct nd_io *ndio_lookup(struct nd_bus *nd_bus, const char *diskname);
bool is_nd_dimm(struct device *dev);
bool is_nd_blk(struct device *dev);
bool is_nd_pmem(struct device *dev);
+#if IS_ENABLED(CONFIG_ND_BTT_DEVS)
+bool is_nd_btt(struct device *dev);
+struct nd_btt *nd_btt_create(struct nd_bus *nd_bus);
+void nd_btt_notify_ndio(struct nd_bus *nd_bus, struct nd_io *ndio);
+#else
+static inline bool is_nd_btt(struct device *dev)
+{
+ return false;
+}
+
+static inline struct nd_btt *nd_btt_create(struct nd_bus *nd_bus)
+{
+ return NULL;
+}
+
+static inline void nd_btt_notify_ndio(struct nd_bus *nd_bus, struct nd_io *ndio)
+{
+}
+#endif
struct nd_bus *walk_to_nd_bus(struct device *nd_dev);
int __init nd_bus_init(void);
void nd_bus_exit(void);
@@ -62,6 +89,7 @@ void nd_synchronize(void);
int nd_bus_register_dimms(struct nd_bus *nd_bus);
int nd_bus_register_regions(struct nd_bus *nd_bus);
int nd_bus_init_interleave_sets(struct nd_bus *nd_bus);
+void __nd_device_register(struct device *dev);
int nd_match_dimm(struct device *dev, void *data);
struct nd_label_id;
char *nd_label_gen_id(struct nd_label_id *label_id, u8 *uuid, u32 flags);
diff --git a/drivers/block/nd/nd.h b/drivers/block/nd/nd.h
index 8eaf0ae82db4..c6ed26d4dcad 100644
--- a/drivers/block/nd/nd.h
+++ b/drivers/block/nd/nd.h
@@ -12,13 +12,19 @@
*/
#ifndef __ND_H__
#define __ND_H__
+#include <linux/genhd.h>
#include <linux/device.h>
#include <linux/mutex.h>
#include <linux/ndctl.h>
#include <linux/types.h>
+#include <linux/fs.h>
#include "libnd.h"
#include "label.h"
+enum {
+ SECTOR_SHIFT = 9,
+};
+
struct nd_dimm_drvdata {
struct device *dev;
int nsindex_size;
@@ -111,6 +117,84 @@ static inline unsigned nd_inc_seq(unsigned seq)
return next[seq & 3];
}
+struct nd_io;
+/**
+ * nd_rw_bytes_fn() - access bytes relative to the "whole disk" namespace device
+ * @ndio: per-namespace context
+ * @buf: source / target for the write / read
+ * @offset: offset relative to the start of the namespace device
+ * @n: num bytes to access
+ * @flags: READ, WRITE, and other REQ_* flags
+ *
+ * Note: Implementations may assume that offset + n never crosses ndio->align
+ */
+typedef int (*nd_rw_bytes_fn)(struct nd_io *ndio, void *buf, size_t offset,
+ size_t n, unsigned long flags);
+#define nd_data_dir(flags) (flags & 1)
+
+/**
+ * struct nd_io - info for byte-aligned access to nd devices
+ * @rw_bytes: operation to perform byte-aligned access
+ * @align: a single ->rw_bytes() request may not cross this alignment
+ * @gendisk: whole disk block device for the namespace
+ * @list: for the core to cache a list of "ndio"s for later association
+ * @dev: namespace device
+ * @claims: list of clients using this interface
+ * @lock: protect @claims mutation
+ */
+struct nd_io {
+ nd_rw_bytes_fn rw_bytes;
+ unsigned long align;
+ struct gendisk *disk;
+ struct list_head list;
+ struct device *dev;
+ struct list_head claims;
+ spinlock_t lock;
+};
+
+struct nd_io_claim;
+typedef void (*ndio_notify_remove_fn)(struct nd_io_claim *ndio_claim);
+
+/**
+ * struct nd_io_claim - instance of a claim on a parent ndio
+ * @notify_remove: ndio is going away, release resources
+ * @holder: object that has claimed this ndio
+ * @parent: ndio in use
+ * @holder: holder device
+ * @list: claim peers
+ *
+ * An ndio may be claimed multiple times, consider the case of a btt
+ * instance per partition on a namespace.
+ */
+struct nd_io_claim {
+ struct nd_io *parent;
+ ndio_notify_remove_fn notify_remove;
+ struct list_head list;
+ struct device *holder;
+};
+
+struct nd_btt {
+ struct device dev;
+ struct nd_io *ndio;
+ struct block_device *backing_dev;
+ unsigned long lbasize;
+ u8 *uuid;
+ u64 offset;
+ int id;
+ struct nd_io_claim *ndio_claim;
+};
+
+static inline u64 nd_partition_offset(struct block_device *bdev)
+{
+ struct hd_struct *p;
+
+ if (bdev == bdev->bd_contains)
+ return 0;
+
+ p = bdev->bd_part;
+ return ((u64) p->start_sect) << SECTOR_SHIFT;
+}
+
enum nd_async_mode {
ND_SYNC,
ND_ASYNC,
@@ -125,12 +209,22 @@ ssize_t nd_sector_size_show(unsigned long current_lbasize,
const unsigned long *supported, char *buf);
ssize_t nd_sector_size_store(struct device *dev, const char *buf,
unsigned long *current_lbasize, const unsigned long *supported);
+int nd_register_ndio(struct nd_io *ndio);
+int nd_unregister_ndio(struct nd_io *ndio);
+void nd_init_ndio(struct nd_io *ndio, nd_rw_bytes_fn rw_bytes,
+ struct device *dev, struct gendisk *disk, unsigned long align);
+void ndio_del_claim(struct nd_io_claim *ndio_claim);
+struct nd_io_claim *ndio_add_claim(struct nd_io *ndio, struct device *holder,
+ ndio_notify_remove_fn notify_remove);
struct nd_dimm;
struct nd_dimm_drvdata *to_ndd(struct nd_mapping *nd_mapping);
int nd_dimm_init_nsarea(struct nd_dimm_drvdata *ndd);
int nd_dimm_init_config_data(struct nd_dimm_drvdata *ndd);
int nd_dimm_set_config_data(struct nd_dimm_drvdata *ndd, size_t offset,
void *buf, size_t len);
+struct nd_btt *to_nd_btt(struct device *dev);
+struct btt_sb;
+u64 nd_btt_sb_checksum(struct btt_sb *btt_sb);
struct nd_region *to_nd_region(struct device *dev);
int nd_region_to_namespace_type(struct nd_region *nd_region);
int nd_region_register_namespaces(struct nd_region *nd_region, int *err);
diff --git a/drivers/block/nd/pmem.c b/drivers/block/nd/pmem.c
index 32e7e74ee25a..900dad61a6b9 100644
--- a/drivers/block/nd/pmem.c
+++ b/drivers/block/nd/pmem.c
@@ -31,6 +31,7 @@
struct pmem_device {
struct request_queue *pmem_queue;
struct gendisk *pmem_disk;
+ struct nd_io ndio;
/* One contiguous memory region per device */
phys_addr_t phys_addr;
@@ -100,6 +101,26 @@ static int pmem_rw_page(struct block_device *bdev, sector_t sector,
return 0;
}
+static int pmem_rw_bytes(struct nd_io *ndio, void *buf, size_t offset,
+ size_t n, unsigned long flags)
+{
+ struct pmem_device *pmem = container_of(ndio, typeof(*pmem), ndio);
+ int rw = nd_data_dir(flags);
+
+ if (unlikely(offset + n > pmem->size)) {
+ dev_WARN_ONCE(ndio->dev, 1, "%s: request out of range\n",
+ __func__);
+ return -EFAULT;
+ }
+
+ if (rw == READ)
+ memcpy(buf, pmem->virt_addr + offset, n);
+ else
+ memcpy(pmem->virt_addr + offset, buf, n);
+
+ return 0;
+}
+
static long pmem_direct_access(struct block_device *bdev, sector_t sector,
void **kaddr, unsigned long *pfn, long size)
{
@@ -179,8 +200,6 @@ static struct pmem_device *pmem_alloc(struct device *dev, struct resource *res)
set_capacity(disk, pmem->size >> 9);
pmem->pmem_disk = disk;
- add_disk(disk);
-
return pmem;
out_free_queue:
@@ -235,7 +254,12 @@ static int nd_pmem_probe(struct device *dev)
if (IS_ERR(pmem))
return PTR_ERR(pmem);
+ nd_bus_lock(dev);
+ add_disk(pmem->pmem_disk);
dev_set_drvdata(dev, pmem);
+ nd_init_ndio(&pmem->ndio, pmem_rw_bytes, dev, pmem->pmem_disk, 0);
+ nd_register_ndio(&pmem->ndio);
+ nd_bus_unlock(dev);
return 0;
}
@@ -244,6 +268,7 @@ static int nd_pmem_remove(struct device *dev)
{
struct pmem_device *pmem = dev_get_drvdata(dev);
+ nd_unregister_ndio(&pmem->ndio);
pmem_free(pmem);
return 0;
}
diff --git a/include/uapi/linux/ndctl.h b/include/uapi/linux/ndctl.h
index 0b4dcabb248a..e595751c613d 100644
--- a/include/uapi/linux/ndctl.h
+++ b/include/uapi/linux/ndctl.h
@@ -181,6 +181,7 @@ static inline const char *nd_dimm_cmd_name(unsigned cmd)
#define ND_DEVICE_NAMESPACE_IO 4 /* legacy persistent memory */
#define ND_DEVICE_NAMESPACE_PMEM 5 /* persistent memory namespace (may alias) */
#define ND_DEVICE_NAMESPACE_BLK 6 /* block-data-window namespace (may alias) */
+#define ND_DEVICE_BTT 7 /* block-translation table device */
enum nd_driver_flags {
ND_DRIVER_DIMM = 1 << ND_DEVICE_DIMM,
@@ -189,6 +190,7 @@ enum nd_driver_flags {
ND_DRIVER_NAMESPACE_IO = 1 << ND_DEVICE_NAMESPACE_IO,
ND_DRIVER_NAMESPACE_PMEM = 1 << ND_DEVICE_NAMESPACE_PMEM,
ND_DRIVER_NAMESPACE_BLK = 1 << ND_DEVICE_NAMESPACE_BLK,
+ ND_DRIVER_BTT = 1 << ND_DEVICE_BTT,
};
enum {
From: Vishal Verma <[email protected]>
BTT stands for Block Translation Table, and is a way to provide power
fail sector atomicity semantics for block devices that have the ability
to perform byte granularity IO. It relies on the ->rw_bytes() capability
of provided nd namespace devices.
The BTT works as a stacked blocked device, and reserves a chunk of space
from the backing device for its accounting metadata. BLK namespaces may
mandate use of a BTT and expect the bus to initialize a BTT if not
already present. Otherwise if a BTT is desired for other namespaces (or
partitions of a namespace) a BTT may be manually configured.
Cc: Andy Lutomirski <[email protected]>
Cc: Boaz Harrosh <[email protected]>
Cc: H. Peter Anvin <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Neil Brown <[email protected]>
Cc: Jeff Moyer <[email protected]>
Cc: Dave Chinner <[email protected]>
Cc: Greg KH <[email protected]>
[jmoyer: fix nmi watchdog timeout in btt_map_init]
[jmoyer: move btt initialization to module load path]
[jmoyer: fix memory leak in the btt initialization path]
[jmoyer: Don't overwrite corrupted arenas]
Signed-off-by: Vishal Verma <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
Documentation/blockdev/btt.txt | 273 ++++++++
drivers/block/nd/Kconfig | 20 +
drivers/block/nd/Makefile | 3
drivers/block/nd/acpi.c | 1
drivers/block/nd/btt.c | 1423 ++++++++++++++++++++++++++++++++++++++++
drivers/block/nd/btt.h | 140 ++++
drivers/block/nd/btt_devs.c | 3
drivers/block/nd/libnd.h | 1
drivers/block/nd/nd-private.h | 1
drivers/block/nd/nd.h | 10
drivers/block/nd/region.c | 67 ++
drivers/block/nd/region_devs.c | 10
12 files changed, 1948 insertions(+), 4 deletions(-)
create mode 100644 Documentation/blockdev/btt.txt
create mode 100644 drivers/block/nd/btt.c
diff --git a/Documentation/blockdev/btt.txt b/Documentation/blockdev/btt.txt
new file mode 100644
index 000000000000..95134d5ec4a0
--- /dev/null
+++ b/Documentation/blockdev/btt.txt
@@ -0,0 +1,273 @@
+BTT - Block Translation Table
+=============================
+
+
+1. Introduction
+---------------
+
+Persistent memory based storage is able to perform IO at byte (or more
+accurately, cache line) granularity. However, we often want to expose such
+storage as traditional block devices. The block drivers for persistent memory
+will do exactly this. However, they do not provide any atomicity guarantees.
+Traditional SSDs typically provide protection against torn sectors in hardware,
+using stored energy in capacitors to complete in-flight block writes, or perhaps
+in firmware. We don't have this luxury with persistent memory - if a write is in
+progress, and we experience a power failure, the block will contain a mix of old
+and new data. Applications may not be prepared to handle such a scenario.
+
+The Block Translation Table (BTT) provides atomic sector update semantics for
+persistent memory devices, so that applications that rely on sector writes not
+being torn can continue to do so. The BTT manifests itself as a stacked block
+device, and reserves a portion of the underlying storage for its metadata. At
+the heart of it, is an indirection table that re-maps all the blocks on the
+volume. It can be thought of as an extremely simple file system that only
+provides atomic sector updates.
+
+
+2. Static Layout
+----------------
+
+The underlying storage on which a BTT can be laid out is not limited in any way.
+The BTT, however, splits the available space into chunks of up to 512 GiB,
+called "Arenas".
+
+Each arena follows the same layout for its metadata, and all references in an
+arena are internal to it (with the exception of one field that points to the
+next arena). The following depicts the "On-disk" metadata layout:
+
+
+ Backing Store +-------> Arena
++---------------+ | +------------------+
+| | | | Arena info block |
+| Arena 0 +---+ | 4K |
+| 512G | +------------------+
+| | | |
++---------------+ | |
+| | | |
+| Arena 1 | | Data Blocks |
+| 512G | | |
+| | | |
++---------------+ | |
+| . | | |
+| . | | |
+| . | | |
+| | | |
+| | | |
++---------------+ +------------------+
+ | |
+ | BTT Map |
+ | |
+ | |
+ +------------------+
+ | |
+ | BTT Flog |
+ | |
+ +------------------+
+ | Info block copy |
+ | 4K |
+ +------------------+
+
+
+3. Theory of Operation
+----------------------
+
+
+a. The BTT Map
+--------------
+
+The map is a simple lookup/indirection table that maps an LBA to an internal
+block. Each map entry is 32 bits. The two most significant bits are special
+flags, and the remaining form the internal block number.
+
+Bit Description
+31 : TRIM flag - marks if the block was trimmed or discarded
+30 : ERROR flag - marks an error block. Cleared on write.
+29 - 0 : Mappings to internal 'postmap' blocks
+
+
+Some of the terminology that will be subsequently used:
+
+External LBA : LBA as made visible to upper layers.
+ABA : Arena Block Address - Block offset/number within an arena
+Premap ABA : The block offset into an arena, which was decided upon by range
+ checking the External LBA
+Postmap ABA : The block number in the "Data Blocks" area obtained after
+ indirection from the map
+nfree : The number of free blocks that are maintained at any given time.
+ This is the number of concurrent writes that can happen to the
+ arena.
+
+
+For example, after adding a BTT, we surface a disk of 1024G. We get a read for
+the external LBA at 768G. This falls into the second arena, and of the 512G
+worth of blocks that this arena contributes, this block is at 256G. Thus, the
+premap ABA is 256G. We now refer to the map, and find out the mapping for block
+'X' (256G) points to block 'Y', say '64'. Thus the postmap ABA is 64.
+
+
+b. The BTT Flog
+---------------
+
+The BTT provides sector atomicity by making every write an "allocating write",
+i.e. Every write goes to a "free" block. A running list of free blocks is
+maintained in the form of the BTT flog. 'Flog' is a combination of the words
+"free list" and "log". The flog contains 'nfree' entries, and an entry contains:
+
+lba : The premap ABA that is being written to
+old_map : The old postmap ABA - after 'this' write completes, this will be a
+ free block.
+new_map : The new postmap ABA. The map will up updated to reflect this
+ lba->postmap_aba mapping, but we log it here in case we have to
+ recover.
+seq : Sequence number to mark which of the 2 sections of this flog entry is
+ valid/newest. It cycles between 01->10->11->01 (binary) under normal
+ operation, with 00 indicating an uninitialized state.
+lba' : alternate lba entry
+old_map': alternate old postmap entry
+new_map': alternate new postmap entry
+seq' : alternate sequence number.
+
+Each of the above fields is 32-bit, making one entry 16 bytes. Flog updates are
+done such that for any entry being written, it:
+a. overwrites the 'old' section in the entry based on sequence numbers
+b. writes the new entry such that the sequence number is written last.
+
+
+c. The concept of lanes
+-----------------------
+
+While 'nfree' describes the number of concurrent IOs an arena can process
+concurrently, 'nlanes' is the number of IOs the BTT device as a whole can
+process.
+ nlanes = min(nfree, num_cpus)
+A lane number is obtained at the start of any IO, and is used for indexing into
+all the on-disk and in-memory data structures for the duration of the IO. It is
+protected by a spinlock.
+
+
+d. In-memory data structure: Read Tracking Table (RTT)
+------------------------------------------------------
+
+Consider a case where we have two threads, one doing reads and the other,
+writes. We can hit a condition where the writer thread grabs a free block to do
+a new IO, but the (slow) reader thread is still reading from it. In other words,
+the reader consulted a map entry, and started reading the corresponding block. A
+writer started writing to the same external LBA, and finished the write updating
+the map for that external LBA to point to its new postmap ABA. At this point the
+internal, postmap block that the reader is (still) reading has been inserted
+into the list of free blocks. If another write comes in for the same LBA, it can
+grab this free block, and start writing to it, causing the reader to read
+incorrect data. To prevent this, we introduce the RTT.
+
+The RTT is a simple, per arena table with 'nfree' entries. Every reader inserts
+into rtt[lane_number], the postmap ABA it is reading, and clears it after the
+read is complete. Every writer thread, after grabbing a free block, checks the
+RTT for its presence. If the postmap free block is in the RTT, it waits till the
+reader clears the RTT entry, and only then starts writing to it.
+
+
+e. In-memory data structure: map locks
+--------------------------------------
+
+Consider a case where two writer threads are writing to the same LBA. There can
+be a race in the following sequence of steps:
+
+free[lane] = map[premap_aba]
+map[premap_aba] = postmap_aba
+
+Both threads can update their respective free[lane] with the same old, freed
+postmap_aba. This has made the layout inconsistent by losing a free entry, and
+at the same time, duplicating another free entry for two lanes.
+
+To solve this, we could have a single map lock (per arena) that has to be taken
+before performing the above sequence, but we feel that could be too contentious.
+Instead we use an array of (nfree) map_locks that is indexed by
+(premap_aba modulo nfree).
+
+
+f. Reconstruction from the Flog
+-------------------------------
+
+On startup, we analyze the BTT flog to create our list of free blocks. We walk
+through all the entries, and for each lane, of the set of two possible
+'sections', we always look at the most recent one only (based on the sequence
+number). The reconstruction rules/steps are simple:
+- Read map[log_entry.lba].
+- If log_entry.new matches the map entry, then log_entry.old is free.
+- If log_entry.new does not match the map entry, then log_entry.new is free.
+ (This case can only be caused by power-fails/unsafe shutdowns)
+
+
+g. Summarizing - Read and Write flows
+-------------------------------------
+
+Read:
+
+1. Convert external LBA to arena number + pre-map ABA
+2. Get a lane (and take lane_lock)
+3. Read map to get the entry for this pre-map ABA
+4. Enter post-map ABA into RTT[lane]
+5. If TRIM flag set in map, return zeroes, and end IO (go to step 8)
+6. If ERROR flag set in map, end IO with EIO (go to step 8)
+7. Read data from this block
+8. Remove post-map ABA entry from RTT[lane]
+9. Release lane (and lane_lock)
+
+Write:
+
+1. Convert external LBA to Arena number + pre-map ABA
+2. Get a lane (and take lane_lock)
+3. Use lane to index into in-memory free list and obtain a new block, next flog
+ index, next sequence number
+4. Scan the RTT to check if free block is present, and spin/wait if it is.
+5. Write data to this free block
+6. Read map to get the existing post-map ABA entry for this pre-map ABA
+7. Write flog entry: [premap_aba / old postmap_aba / new postmap_aba / seq_num]
+8. Write new post-map ABA into map.
+9. Write old post-map entry into the free list
+10. Calculate next sequence number and write into the free list entry
+11. Release lane (and lane_lock)
+
+
+4. Error Handling
+=================
+
+An arena would be in an error state if any of the metadata is corrupted
+irrecoverably, either due to a bug or a media error. The following conditions
+indicate an error:
+- Info block checksum does not match (and recovering from the copy also fails)
+- All internal available blocks are not uniquely and entirely addressed by the
+ sum of mapped blocks and free blocks (from the BTT flog).
+- Rebuilding free list from the flog reveals missing/duplicate/impossible
+ entries
+- A map entry is out of bounds
+
+If any of these error conditions are encountered, the arena is put into a read
+only state using a flag in the info block.
+
+
+5. In-kernel usage
+==================
+
+Any block driver that supports byte granularity IO to the storage may register
+with the BTT. It will have to provide the rw_bytes interface in its
+block_device_operations struct:
+
+ int (*rw_bytes)(struct gendisk *, void *, size_t, off_t, int rw);
+
+It may register with the BTT after it adds its own gendisk, using btt_init:
+
+ struct btt *btt_init(struct gendisk *disk, unsigned long long rawsize,
+ u32 lbasize, u8 uuid[], int maxlane);
+
+note that maxlane is the maximum amount of concurrency the driver wishes to
+allow the BTT to use.
+
+The BTT 'disk' appears as a stacked block device that grabs the underlying block
+device in the O_EXCL mode.
+
+When the driver wishes to remove the backing disk, it should similarly call
+btt_fini using the same struct btt* handle that was provided to it by btt_init.
+
+ void btt_fini(struct btt *btt);
+
diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
index 15896db4de37..612bf2b14283 100644
--- a/drivers/block/nd/Kconfig
+++ b/drivers/block/nd/Kconfig
@@ -93,9 +93,25 @@ config BLK_DEV_PMEM
capable of DAX (direct-access) file system mappings. See
Documentation/blockdev/nd.txt for more details.
- Say Y if you want to use a NVDIMM described by NFIT
+ Say Y if you want to use a NVDIMM described by ACPI, E820, etc...
config ND_BTT_DEVS
- def_bool y
+ bool
+
+config ND_BTT
+ tristate "BTT: Block Translation Table (atomic sector updates)"
+ depends on LIBND
+ default LIBND
+ select ND_BTT_DEVS
+
+config ND_MAX_REGIONS
+ int "Maximum number of regions supported by the sub-system"
+ default 64
+ ---help---
+ A 'region' corresponds to an individual DIMM or an interleave
+ set of DIMMs. A typical maximally configured system may have
+ up to 32 DIMMs.
+
+ Leave the default of 64 if you are unsure.
endif
diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
index 0c6d64b7a69d..7d778b4523d4 100644
--- a/drivers/block/nd/Makefile
+++ b/drivers/block/nd/Makefile
@@ -17,6 +17,7 @@ obj-$(CONFIG_ND_ACPI) += nd_acpi.o
obj-$(CONFIG_ND_E820) += nd_e820.o
obj-$(CONFIG_NFIT_TEST) += test/
obj-$(CONFIG_BLK_DEV_PMEM) += nd_pmem.o
+obj-$(CONFIG_ND_BTT) += nd_btt.o
nd_acpi-y := acpi.o
@@ -24,6 +25,8 @@ nd_e820-y := e820.o
nd_pmem-y := pmem.o
+nd_btt-y := btt.o
+
libnd-y := core.o
libnd-y += bus.o
libnd-y += dimm_devs.o
diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
index d34cefe38e2f..5b9997fbc344 100644
--- a/drivers/block/nd/acpi.c
+++ b/drivers/block/nd/acpi.c
@@ -926,6 +926,7 @@ static int nd_acpi_register_region(struct acpi_nfit_desc *acpi_desc,
} else {
nd_mapping->size = nfit_mem->bdw->blk_capacity;
nd_mapping->start = nfit_mem->bdw->blk_offset;
+ ndr_desc.num_lanes = nfit_mem->bdw->num_bdw;
}
ndr_desc.nd_mapping = nd_mapping;
diff --git a/drivers/block/nd/btt.c b/drivers/block/nd/btt.c
new file mode 100644
index 000000000000..abcefb7aeed1
--- /dev/null
+++ b/drivers/block/nd/btt.c
@@ -0,0 +1,1423 @@
+/*
+ * Block Translation Table
+ * Copyright (c) 2014-2015, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+#include <linux/highmem.h>
+#include <linux/debugfs.h>
+#include <linux/blkdev.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/mutex.h>
+#include <linux/hdreg.h>
+#include <linux/genhd.h>
+#include <linux/sizes.h>
+#include <linux/ndctl.h>
+#include <linux/fs.h>
+#include <linux/nd.h>
+#include "btt.h"
+#include "nd.h"
+
+enum log_ent_request {
+ LOG_NEW_ENT = 0,
+ LOG_OLD_ENT
+};
+
+static int btt_major;
+
+static int nd_btt_rw_bytes(struct nd_btt *nd_btt, void *buf, size_t offset,
+ size_t n, unsigned long flags)
+{
+ struct nd_io *ndio = nd_btt->ndio;
+
+ if (unlikely(nd_data_dir(flags) == WRITE)
+ && bdev_read_only(nd_btt->backing_dev))
+ return -EACCES;
+
+ return ndio->rw_bytes(ndio, buf, offset + nd_btt->offset, n, flags);
+}
+
+static int arena_rw_bytes(struct arena_info *arena, void *buf, size_t n,
+ size_t offset, unsigned long flags)
+{
+ /* yes, FIXME, 'offset' and 'n' are swapped */
+ return nd_btt_rw_bytes(arena->nd_btt, buf, offset, n, flags);
+}
+
+static int btt_info_write(struct arena_info *arena, struct btt_sb *super)
+{
+ int ret;
+
+ ret = arena_rw_bytes(arena, super, sizeof(struct btt_sb),
+ arena->info2off, WRITE);
+ if (ret)
+ return ret;
+
+ return arena_rw_bytes(arena, super, sizeof(struct btt_sb),
+ arena->infooff, WRITE);
+}
+
+static int btt_info_read(struct arena_info *arena, struct btt_sb *super)
+{
+ WARN_ON(!super);
+ return arena_rw_bytes(arena, super, sizeof(struct btt_sb),
+ arena->infooff, READ);
+
+}
+
+/*
+ * 'raw' version of btt_map write
+ * Assumptions:
+ * mapping is in little-endian
+ * mapping contains 'E' and 'Z' flags as desired
+ */
+static int __btt_map_write(struct arena_info *arena, u32 lba, __le32 mapping)
+{
+ u64 ns_off = arena->mapoff + (lba * MAP_ENT_SIZE);
+
+ WARN_ON(lba >= arena->external_nlba);
+ return arena_rw_bytes(arena, &mapping, MAP_ENT_SIZE, ns_off, WRITE);
+}
+
+static int btt_map_write(struct arena_info *arena, u32 lba, u32 mapping,
+ u32 z_flag, u32 e_flag)
+{
+ u32 ze;
+ __le32 mapping_le;
+
+ /*
+ * This 'mapping' is supposed to be just the LBA mapping, without
+ * any flags set, so strip the flag bits.
+ */
+ mapping &= MAP_LBA_MASK;
+
+ ze = (z_flag << 1) + e_flag;
+ switch (ze) {
+ case 0:
+ /*
+ * We want to set neither of the Z or E flags, and
+ * in the actual layout, this means setting the bit
+ * positions of both to '1' to indicate a 'normal'
+ * map entry
+ */
+ mapping |= MAP_ENT_NORMAL;
+ break;
+ case 1:
+ mapping |= (1 << MAP_ERR_SHIFT);
+ break;
+ case 2:
+ mapping |= (1 << MAP_TRIM_SHIFT);
+ break;
+ default:
+ /*
+ * The case where Z and E are both sent in as '1' could be
+ * construed as a valid 'normal' case, but we decide not to,
+ * to avoid confusion
+ */
+ WARN_ONCE(1, "Invalid use of Z and E flags\n");
+ return -EIO;
+ }
+
+ mapping_le = cpu_to_le32(mapping);
+ return __btt_map_write(arena, lba, mapping_le);
+}
+
+static int btt_map_read(struct arena_info *arena, u32 lba, u32 *mapping,
+ int *trim, int *error)
+{
+ int ret;
+ __le32 in;
+ u32 raw_mapping, postmap, ze, z_flag, e_flag;
+ u64 ns_off = arena->mapoff + (lba * MAP_ENT_SIZE);
+
+ WARN_ON(lba >= arena->external_nlba);
+
+ ret = arena_rw_bytes(arena, &in, MAP_ENT_SIZE, ns_off, READ);
+ if (ret)
+ return ret;
+
+ raw_mapping = le32_to_cpu(in);
+
+ z_flag = (raw_mapping & MAP_TRIM_MASK) >> MAP_TRIM_SHIFT;
+ e_flag = (raw_mapping & MAP_ERR_MASK) >> MAP_ERR_SHIFT;
+ ze = (z_flag << 1) + e_flag;
+ postmap = raw_mapping & MAP_LBA_MASK;
+
+ /* Reuse the {z,e}_flag variables for *trim and *error */
+ z_flag = 0;
+ e_flag = 0;
+
+ switch (ze) {
+ case 0:
+ /* Initial state. Return postmap = premap */
+ *mapping = lba;
+ break;
+ case 1:
+ *mapping = postmap;
+ e_flag = 1;
+ break;
+ case 2:
+ *mapping = postmap;
+ z_flag = 1;
+ break;
+ case 3:
+ *mapping = postmap;
+ break;
+ default:
+ return -EIO;
+ }
+
+ if (trim)
+ *trim = z_flag;
+ if (error)
+ *error = e_flag;
+
+ return ret;
+}
+
+static int btt_log_read_pair(struct arena_info *arena, u32 lane,
+ struct log_entry *ent)
+{
+ WARN_ON(!ent);
+ return arena_rw_bytes(arena, ent, 2 * LOG_ENT_SIZE,
+ arena->logoff + (2 * lane * LOG_ENT_SIZE), READ);
+}
+
+static struct dentry *debugfs_root;
+
+static void arena_debugfs_init(struct arena_info *a, struct dentry *parent,
+ int idx)
+{
+ char dirname[32];
+ struct dentry *d;
+
+ /* If for some reason, parent bttN was not created, exit */
+ if (!parent)
+ return;
+
+ snprintf(dirname, 32, "arena%d", idx);
+ d = debugfs_create_dir(dirname, parent);
+ if (IS_ERR_OR_NULL(d))
+ return;
+ a->debugfs_dir = d;
+
+ debugfs_create_x64("size", S_IRUGO, d, &a->size);
+ debugfs_create_x64("external_lba_start", S_IRUGO, d,
+ &a->external_lba_start);
+ debugfs_create_x32("internal_nlba", S_IRUGO, d, &a->internal_nlba);
+ debugfs_create_u32("internal_lbasize", S_IRUGO, d,
+ &a->internal_lbasize);
+ debugfs_create_x32("external_nlba", S_IRUGO, d, &a->external_nlba);
+ debugfs_create_u32("external_lbasize", S_IRUGO, d,
+ &a->external_lbasize);
+ debugfs_create_u32("nfree", S_IRUGO, d, &a->nfree);
+ debugfs_create_u16("version_major", S_IRUGO, d, &a->version_major);
+ debugfs_create_u16("version_minor", S_IRUGO, d, &a->version_minor);
+ debugfs_create_x64("nextoff", S_IRUGO, d, &a->nextoff);
+ debugfs_create_x64("infooff", S_IRUGO, d, &a->infooff);
+ debugfs_create_x64("dataoff", S_IRUGO, d, &a->dataoff);
+ debugfs_create_x64("mapoff", S_IRUGO, d, &a->mapoff);
+ debugfs_create_x64("logoff", S_IRUGO, d, &a->logoff);
+ debugfs_create_x64("info2off", S_IRUGO, d, &a->info2off);
+ debugfs_create_x32("flags", S_IRUGO, d, &a->flags);
+}
+
+static void btt_debugfs_init(struct btt *btt)
+{
+ int i = 0;
+ struct arena_info *arena;
+
+ btt->debugfs_dir = debugfs_create_dir(dev_name(&btt->nd_btt->dev),
+ debugfs_root);
+ if (IS_ERR_OR_NULL(btt->debugfs_dir))
+ return;
+
+ list_for_each_entry(arena, &btt->arena_list, list) {
+ arena_debugfs_init(arena, btt->debugfs_dir, i);
+ i++;
+ }
+}
+
+/*
+ * This function accepts two log entries, and uses the
+ * sequence number to find the 'older' entry.
+ * It also updates the sequence number in this old entry to
+ * make it the 'new' one if the mark_flag is set.
+ * Finally, it returns which of the entries was the older one.
+ *
+ * TODO The logic feels a bit kludge-y. make it better..
+ */
+static int btt_log_get_old(struct log_entry *ent)
+{
+ int old;
+
+ /*
+ * the first ever time this is seen, the entry goes into [0]
+ * the next time, the following logic works out to put this
+ * (next) entry into [1]
+ */
+ if (ent[0].seq == 0) {
+ ent[0].seq = cpu_to_le32(1);
+ return 0;
+ }
+
+ if (ent[0].seq == ent[1].seq)
+ return -EINVAL;
+ if (le32_to_cpu(ent[0].seq) + le32_to_cpu(ent[1].seq) > 5)
+ return -EINVAL;
+
+ if (le32_to_cpu(ent[0].seq) < le32_to_cpu(ent[1].seq)) {
+ if (le32_to_cpu(ent[1].seq) - le32_to_cpu(ent[0].seq) == 1)
+ old = 0;
+ else
+ old = 1;
+ } else {
+ if (le32_to_cpu(ent[0].seq) - le32_to_cpu(ent[1].seq) == 1)
+ old = 1;
+ else
+ old = 0;
+ }
+
+ return old;
+}
+
+static struct device *to_dev(struct arena_info *arena)
+{
+ return &arena->nd_btt->dev;
+}
+
+/*
+ * This function copies the desired (old/new) log entry into ent if
+ * it is not NULL. It returns the sub-slot number (0 or 1)
+ * where the desired log entry was found. Negative return values
+ * indicate errors.
+ */
+static int btt_log_read(struct arena_info *arena, u32 lane,
+ struct log_entry *ent, int old_flag)
+{
+ int ret;
+ int old_ent, ret_ent;
+ struct log_entry log[2];
+
+ ret = btt_log_read_pair(arena, lane, log);
+ if (ret)
+ return -EIO;
+
+ old_ent = btt_log_get_old(log);
+ if (old_ent < 0 || old_ent > 1) {
+ dev_info(to_dev(arena),
+ "log corruption (%d): lane %d seq [%d, %d]\n",
+ old_ent, lane, log[0].seq, log[1].seq);
+ /* TODO set error state? */
+ return -EIO;
+ }
+
+ ret_ent = (old_flag ? old_ent : (1 - old_ent));
+
+ if (ent != NULL)
+ memcpy(ent, &log[ret_ent], LOG_ENT_SIZE);
+
+ return ret_ent;
+}
+
+/*
+ * This function commits a log entry to media
+ * It does _not_ prepare the freelist entry for the next write
+ * btt_flog_write is the wrapper for updating the freelist elements
+ */
+static int __btt_log_write(struct arena_info *arena, u32 lane,
+ u32 sub, struct log_entry *ent)
+{
+ int ret;
+ /*
+ * Ignore the padding in log_entry for calculating log_half.
+ * The entry is 'committed' when we write the sequence number,
+ * and we want to ensure that that is the last thing written.
+ * We don't bother writing the padding as that would be extra
+ * media wear and write amplification
+ */
+ unsigned int log_half = (LOG_ENT_SIZE - 2 * sizeof(u64)) / 2;
+ u64 ns_off = arena->logoff + (((2 * lane) + sub) * LOG_ENT_SIZE);
+ void *src = ent;
+
+ /* split the 16B write into atomic, durable halves */
+ ret = arena_rw_bytes(arena, src, log_half, ns_off, WRITE);
+ if (ret)
+ return ret;
+
+ ns_off += log_half;
+ src += log_half;
+ return arena_rw_bytes(arena, src, log_half, ns_off, WRITE);
+}
+
+static int btt_flog_write(struct arena_info *arena, u32 lane, u32 sub,
+ struct log_entry *ent)
+{
+ int ret;
+
+ ret = __btt_log_write(arena, lane, sub, ent);
+ if (ret)
+ return ret;
+
+ /* prepare the next free entry */
+ arena->freelist[lane].sub = 1 - arena->freelist[lane].sub;
+ if (++(arena->freelist[lane].seq) == 4)
+ arena->freelist[lane].seq = 1;
+ arena->freelist[lane].block = le32_to_cpu(ent->old_map);
+
+ return ret;
+}
+
+/*
+ * This function initializes the BTT map to a state with all externally
+ * exposed blocks having an identity mapping, and the TRIM flag set
+ */
+static int btt_map_init(struct arena_info *arena)
+{
+ int ret = -EINVAL;
+ void *zerobuf;
+ size_t offset = 0;
+ size_t chunk_size = SZ_2M;
+ size_t mapsize = arena->logoff - arena->mapoff;
+
+ zerobuf = kzalloc(chunk_size, GFP_KERNEL);
+ if (!zerobuf)
+ return -ENOMEM;
+
+ while (mapsize) {
+ size_t size = min(mapsize, chunk_size);
+
+ ret = arena_rw_bytes(arena, zerobuf, size,
+ arena->mapoff + offset, WRITE);
+ if (ret)
+ goto free;
+
+ offset += size;
+ mapsize -= size;
+ cond_resched();
+ }
+
+ free:
+ kfree(zerobuf);
+ return ret;
+}
+
+/*
+ * This function initializes the BTT log with 'fake' entries pointing
+ * to the initial reserved set of blocks as being free
+ */
+static int btt_log_init(struct arena_info *arena)
+{
+ int ret;
+ u32 i;
+ struct log_entry log, zerolog;
+
+ memset(&zerolog, 0, sizeof(zerolog));
+
+ for (i = 0; i < arena->nfree; i++) {
+ log.lba = cpu_to_le32(i);
+ log.old_map = cpu_to_le32(arena->external_nlba + i);
+ log.new_map = cpu_to_le32(arena->external_nlba + i);
+ log.seq = cpu_to_le32(LOG_SEQ_INIT);
+ ret = __btt_log_write(arena, i, 0, &log);
+ if (ret)
+ return ret;
+ ret = __btt_log_write(arena, i, 1, &zerolog);
+ if (ret)
+ return ret;
+ }
+
+ return 0;
+}
+
+static int btt_freelist_init(struct arena_info *arena)
+{
+ int old, new, ret;
+ u32 i, map_entry;
+ struct log_entry log_new, log_old;
+
+ arena->freelist = kcalloc(arena->nfree, sizeof(struct free_entry),
+ GFP_KERNEL);
+ if (!arena->freelist)
+ return -ENOMEM;
+
+ for (i = 0; i < arena->nfree; i++) {
+ old = btt_log_read(arena, i, &log_old, LOG_OLD_ENT);
+ if (old < 0)
+ return old;
+
+ new = btt_log_read(arena, i, &log_new, LOG_NEW_ENT);
+ if (new < 0)
+ return new;
+
+ /* sub points to the next one to be overwritten */
+ arena->freelist[i].sub = 1 - new;
+ arena->freelist[i].seq = nd_inc_seq(le32_to_cpu(log_new.seq));
+ arena->freelist[i].block = le32_to_cpu(log_new.old_map);
+
+ /* This implies a newly created or untouched flog entry */
+ if (log_new.old_map == log_new.new_map)
+ continue;
+
+ /* Check if map recovery is needed */
+ ret = btt_map_read(arena, le32_to_cpu(log_new.lba), &map_entry,
+ NULL, NULL);
+ if (ret)
+ return ret;
+ if ((le32_to_cpu(log_new.new_map) != map_entry) &&
+ (le32_to_cpu(log_new.old_map) == map_entry)) {
+ /*
+ * Last transaction wrote the flog, but wasn't able
+ * to complete the map write. So fix up the map.
+ */
+ ret = btt_map_write(arena, le32_to_cpu(log_new.lba),
+ le32_to_cpu(log_new.new_map), 0, 0);
+ if (ret)
+ return ret;
+ }
+
+ }
+
+ return 0;
+}
+
+static int btt_rtt_init(struct arena_info *arena)
+{
+ arena->rtt = kcalloc(arena->nfree, sizeof(u32), GFP_KERNEL);
+ if (arena->rtt == NULL)
+ return -ENOMEM;
+
+ return 0;
+}
+
+static int btt_maplocks_init(struct arena_info *arena)
+{
+ u32 i;
+
+ arena->map_locks = kcalloc(arena->nfree, sizeof(struct aligned_lock),
+ GFP_KERNEL);
+ if (!arena->map_locks)
+ return -ENOMEM;
+
+ for (i = 0; i < arena->nfree; i++)
+ spin_lock_init(&arena->map_locks[i].lock);
+
+ return 0;
+}
+
+static struct arena_info *alloc_arena(struct btt *btt, size_t size,
+ size_t start, size_t arena_off)
+{
+ struct arena_info *arena;
+ u64 logsize, mapsize, datasize;
+ u64 available = size;
+
+ arena = kzalloc(sizeof(struct arena_info), GFP_KERNEL);
+ if (!arena)
+ return NULL;
+ arena->nd_btt = btt->nd_btt;
+
+ if (!size)
+ return arena;
+
+ arena->size = size;
+ arena->external_lba_start = start;
+ arena->external_lbasize = btt->lbasize;
+ arena->internal_lbasize = roundup(arena->external_lbasize,
+ INT_LBASIZE_ALIGNMENT);
+ arena->nfree = BTT_DEFAULT_NFREE;
+ arena->version_major = 1;
+ arena->version_minor = 1;
+
+ if (available % BTT_PG_SIZE)
+ available -= (available % BTT_PG_SIZE);
+
+ /* Two pages are reserved for the super block and its copy */
+ available -= 2 * BTT_PG_SIZE;
+
+ /* The log takes a fixed amount of space based on nfree */
+ logsize = roundup(2 * arena->nfree * sizeof(struct log_entry),
+ BTT_PG_SIZE);
+ available -= logsize;
+
+ /* Calculate optimal split between map and data area */
+ arena->internal_nlba = div_u64(available - BTT_PG_SIZE,
+ arena->internal_lbasize + MAP_ENT_SIZE);
+ arena->external_nlba = arena->internal_nlba - arena->nfree;
+
+ mapsize = roundup((arena->external_nlba * MAP_ENT_SIZE), BTT_PG_SIZE);
+ datasize = available - mapsize;
+
+ /* 'Absolute' values, relative to start of storage space */
+ arena->infooff = arena_off;
+ arena->dataoff = arena->infooff + BTT_PG_SIZE;
+ arena->mapoff = arena->dataoff + datasize;
+ arena->logoff = arena->mapoff + mapsize;
+ arena->info2off = arena->logoff + logsize;
+ return arena;
+}
+
+static void free_arenas(struct btt *btt)
+{
+ struct arena_info *arena, *next;
+
+ list_for_each_entry_safe(arena, next, &btt->arena_list, list) {
+ list_del(&arena->list);
+ kfree(arena->rtt);
+ kfree(arena->map_locks);
+ kfree(arena->freelist);
+ debugfs_remove_recursive(arena->debugfs_dir);
+ kfree(arena);
+ }
+}
+
+/*
+ * This function checks if the metadata layout is valid and error free
+ */
+static int arena_is_valid(struct arena_info *arena, struct btt_sb *super,
+ u8 *uuid)
+{
+ u64 checksum;
+
+ if (memcmp(super->uuid, uuid, 16))
+ return 0;
+
+ checksum = le64_to_cpu(super->checksum);
+ super->checksum = 0;
+ if (checksum != nd_btt_sb_checksum(super))
+ return 0;
+ super->checksum = cpu_to_le64(checksum);
+
+ /* TODO: figure out action for this */
+ if ((le32_to_cpu(super->flags) & IB_FLAG_ERROR_MASK) != 0)
+ dev_info(to_dev(arena), "Found arena with an error flag\n");
+
+ return 1;
+}
+
+/*
+ * This function reads an existing valid btt superblock and
+ * populates the corresponding arena_info struct
+ */
+static void parse_arena_meta(struct arena_info *arena, struct btt_sb *super,
+ u64 arena_off)
+{
+ arena->internal_nlba = le32_to_cpu(super->internal_nlba);
+ arena->internal_lbasize = le32_to_cpu(super->internal_lbasize);
+ arena->external_nlba = le32_to_cpu(super->external_nlba);
+ arena->external_lbasize = le32_to_cpu(super->external_lbasize);
+ arena->nfree = le32_to_cpu(super->nfree);
+ arena->version_major = le16_to_cpu(super->version_major);
+ arena->version_minor = le16_to_cpu(super->version_minor);
+
+ arena->nextoff = (super->nextoff == 0) ? 0 : (arena_off +
+ le64_to_cpu(super->nextoff));
+ arena->infooff = arena_off;
+ arena->dataoff = arena_off + le64_to_cpu(super->dataoff);
+ arena->mapoff = arena_off + le64_to_cpu(super->mapoff);
+ arena->logoff = arena_off + le64_to_cpu(super->logoff);
+ arena->info2off = arena_off + le64_to_cpu(super->info2off);
+
+ arena->size = (super->nextoff > 0) ? (le64_to_cpu(super->nextoff)) :
+ (arena->info2off - arena->infooff + BTT_PG_SIZE);
+
+ arena->flags = le32_to_cpu(super->flags);
+}
+
+static int discover_arenas(struct btt *btt)
+{
+ int ret = 0;
+ struct arena_info *arena;
+ struct btt_sb *super;
+ size_t remaining = btt->rawsize;
+ u64 cur_nlba = 0;
+ size_t cur_off = 0;
+ int num_arenas = 0;
+
+ super = kzalloc(sizeof(*super), GFP_KERNEL);
+ if (!super)
+ return -ENOMEM;
+
+ while (remaining) {
+ /* Alloc memory for arena */
+ arena = alloc_arena(btt, 0, 0, 0);
+ if (!arena) {
+ ret = -ENOMEM;
+ goto out_super;
+ }
+
+ arena->infooff = cur_off;
+ ret = btt_info_read(arena, super);
+ if (ret)
+ goto out;
+
+ if (!arena_is_valid(arena, super, btt->nd_btt->uuid)) {
+ if (remaining == btt->rawsize) {
+ btt->init_state = INIT_NOTFOUND;
+ dev_info(to_dev(arena), "No existing arenas\n");
+ goto out;
+ } else {
+ dev_info(to_dev(arena),
+ "Found corrupted metadata!\n");
+ ret = -ENODEV;
+ goto out;
+ }
+ }
+
+ arena->external_lba_start = cur_nlba;
+ parse_arena_meta(arena, super, cur_off);
+
+ ret = btt_freelist_init(arena);
+ if (ret)
+ goto out;
+
+ ret = btt_rtt_init(arena);
+ if (ret)
+ goto out;
+
+ ret = btt_maplocks_init(arena);
+ if (ret)
+ goto out;
+
+ list_add_tail(&arena->list, &btt->arena_list);
+
+ remaining -= arena->size;
+ cur_off += arena->size;
+ cur_nlba += arena->external_nlba;
+ num_arenas++;
+
+ if (arena->nextoff == 0)
+ break;
+ }
+ btt->num_arenas = num_arenas;
+ btt->nlba = cur_nlba;
+ btt->init_state = INIT_READY;
+
+ kfree(super);
+ return ret;
+
+ out:
+ kfree(arena);
+ free_arenas(btt);
+ out_super:
+ kfree(super);
+ return ret;
+}
+
+static int create_arenas(struct btt *btt)
+{
+ size_t remaining = btt->rawsize;
+ size_t cur_off = 0;
+
+ while (remaining) {
+ struct arena_info *arena;
+ size_t arena_size = min_t(u64, ARENA_MAX_SIZE, remaining);
+
+ remaining -= arena_size;
+ if (arena_size < ARENA_MIN_SIZE)
+ break;
+
+ arena = alloc_arena(btt, arena_size, btt->nlba, cur_off);
+ if (!arena) {
+ free_arenas(btt);
+ return -ENOMEM;
+ }
+ btt->nlba += arena->external_nlba;
+ if (remaining >= ARENA_MIN_SIZE)
+ arena->nextoff = arena->size;
+ else
+ arena->nextoff = 0;
+ cur_off += arena_size;
+ list_add_tail(&arena->list, &btt->arena_list);
+ }
+
+ return 0;
+}
+
+/*
+ * This function completes arena initialization by writing
+ * all the metadata.
+ * It is only called for an uninitialized arena when a write
+ * to that arena occurs for the first time.
+ */
+static int btt_arena_write_layout(struct arena_info *arena, u8 *uuid)
+{
+ int ret;
+ struct btt_sb *super;
+
+ ret = btt_map_init(arena);
+ if (ret)
+ return ret;
+
+ ret = btt_log_init(arena);
+ if (ret)
+ return ret;
+
+ super = kzalloc(sizeof(struct btt_sb), GFP_NOIO);
+ if (!super)
+ return -ENOMEM;
+
+ strncpy(super->signature, BTT_SIG, BTT_SIG_LEN);
+ memcpy(super->uuid, uuid, 16);
+ super->flags = cpu_to_le32(arena->flags);
+ super->version_major = cpu_to_le16(arena->version_major);
+ super->version_minor = cpu_to_le16(arena->version_minor);
+ super->external_lbasize = cpu_to_le32(arena->external_lbasize);
+ super->external_nlba = cpu_to_le32(arena->external_nlba);
+ super->internal_lbasize = cpu_to_le32(arena->internal_lbasize);
+ super->internal_nlba = cpu_to_le32(arena->internal_nlba);
+ super->nfree = cpu_to_le32(arena->nfree);
+ super->infosize = cpu_to_le32(sizeof(struct btt_sb));
+
+ /* TODO: make these relative to arena start. For now we get this
+ * since each file = 1 arena = 1 dimm, but will change */
+ super->nextoff = cpu_to_le64(arena->nextoff);
+ /*
+ * Subtract arena->infooff (arena start) so numbers are relative
+ * to 'this' arena
+ */
+ super->dataoff = cpu_to_le64(arena->dataoff - arena->infooff);
+ super->mapoff = cpu_to_le64(arena->mapoff - arena->infooff);
+ super->logoff = cpu_to_le64(arena->logoff - arena->infooff);
+ super->info2off = cpu_to_le64(arena->info2off - arena->infooff);
+
+ super->flags = 0;
+ super->checksum = cpu_to_le64(nd_btt_sb_checksum(super));
+
+ ret = btt_info_write(arena, super);
+
+ kfree(super);
+ return ret;
+}
+
+/*
+ * This function completes the initialization for the BTT namespace
+ * such that it is ready to accept IOs
+ */
+static int btt_meta_init(struct btt *btt)
+{
+ int ret = 0;
+ struct arena_info *arena;
+
+ mutex_lock(&btt->init_lock);
+ list_for_each_entry(arena, &btt->arena_list, list) {
+ ret = btt_arena_write_layout(arena, btt->nd_btt->uuid);
+ if (ret)
+ goto unlock;
+
+ ret = btt_freelist_init(arena);
+ if (ret)
+ goto unlock;
+
+ ret = btt_rtt_init(arena);
+ if (ret)
+ goto unlock;
+
+ ret = btt_maplocks_init(arena);
+ if (ret)
+ goto unlock;
+ }
+
+ btt->init_state = INIT_READY;
+
+ unlock:
+ mutex_unlock(&btt->init_lock);
+ return ret;
+}
+
+/*
+ * This function calculates the arena in which the given LBA lies
+ * by doing a linear walk. This is acceptable since we expect only
+ * a few arenas. If we have backing devices that get much larger,
+ * we can construct a balanced binary tree of arenas at init time
+ * so that this range search becomes faster.
+ */
+static int lba_to_arena(struct btt *btt, sector_t sector, __u32 *premap,
+ struct arena_info **arena)
+{
+ struct arena_info *arena_list;
+ __u64 lba = div_u64(sector << SECTOR_SHIFT, btt->lbasize);
+
+ list_for_each_entry(arena_list, &btt->arena_list, list) {
+ if (lba < arena_list->external_nlba) {
+ *arena = arena_list;
+ *premap = lba;
+ return 0;
+ }
+ lba -= arena_list->external_nlba;
+ }
+
+ return -EIO;
+}
+
+/*
+ * The following (lock_map, unlock_map) are mostly just to improve
+ * readability, since they index into an array of locks
+ */
+static void lock_map(struct arena_info *arena, u32 premap)
+{
+ u32 idx = (premap * MAP_ENT_SIZE / L1_CACHE_BYTES) % arena->nfree;
+
+ spin_lock(&arena->map_locks[idx].lock);
+}
+
+static void unlock_map(struct arena_info *arena, u32 premap)
+{
+ u32 idx = (premap * MAP_ENT_SIZE / L1_CACHE_BYTES) % arena->nfree;
+
+ spin_unlock(&arena->map_locks[idx].lock);
+}
+
+static u64 to_namespace_offset(struct arena_info *arena, u64 lba)
+{
+ return arena->dataoff + ((u64)lba * arena->internal_lbasize);
+}
+
+static int btt_data_read(struct arena_info *arena, struct page *page,
+ unsigned int off, u32 lba, u32 len)
+{
+ int ret;
+ u64 nsoff = to_namespace_offset(arena, lba);
+ void *mem = kmap_atomic(page);
+
+ ret = arena_rw_bytes(arena, mem + off, len, nsoff, READ);
+ kunmap_atomic(mem);
+
+ return ret;
+}
+
+static int btt_data_write(struct arena_info *arena, u32 lba,
+ struct page *page, unsigned int off, u32 len)
+{
+ int ret;
+ u64 nsoff = to_namespace_offset(arena, lba);
+ void *mem = kmap_atomic(page);
+
+ ret = arena_rw_bytes(arena, mem + off, len, nsoff, WRITE);
+ kunmap_atomic(mem);
+
+ return ret;
+}
+
+static void zero_fill_data(struct page *page, unsigned int off, u32 len)
+{
+ void *mem = kmap_atomic(page);
+
+ memset(mem + off, 0, len);
+ kunmap_atomic(mem);
+}
+
+static int btt_read_pg(struct btt *btt, struct page *page, unsigned int off,
+ sector_t sector, unsigned int len)
+{
+ int ret = 0;
+ int t_flag, e_flag;
+ struct arena_info *arena = NULL;
+ u32 lane = 0, premap, postmap;
+
+ while (len) {
+ u32 cur_len;
+
+ lane = nd_region_acquire_lane(btt->nd_region);
+
+ ret = lba_to_arena(btt, sector, &premap, &arena);
+ if (ret)
+ goto out_lane;
+
+ cur_len = min(arena->external_lbasize, len);
+
+ ret = btt_map_read(arena, premap, &postmap, &t_flag, &e_flag);
+ if (ret)
+ goto out_lane;
+
+ /*
+ * We loop to make sure that the post map LBA didn't change
+ * from under us between writing the RTT and doing the actual
+ * read.
+ */
+ while (1) {
+ u32 new_map;
+
+ if (t_flag) {
+ zero_fill_data(page, off, cur_len);
+ goto out_lane;
+ }
+
+ if (e_flag) {
+ ret = -EIO;
+ goto out_lane;
+ }
+
+ arena->rtt[lane] = RTT_VALID | postmap;
+ /*
+ * Barrier to make sure this write is not reordered
+ * to do the verification map_read before the RTT store
+ */
+ barrier();
+
+ ret = btt_map_read(arena, premap, &new_map, &t_flag,
+ &e_flag);
+ if (ret)
+ goto out_rtt;
+
+ if (postmap == new_map)
+ break;
+
+ postmap = new_map;
+ }
+
+ ret = btt_data_read(arena, page, off, postmap, cur_len);
+ if (ret)
+ goto out_rtt;
+
+ arena->rtt[lane] = RTT_INVALID;
+ nd_region_release_lane(btt->nd_region, lane);
+
+ len -= cur_len;
+ off += cur_len;
+ sector += arena->external_lbasize >> SECTOR_SHIFT;
+ }
+
+ return 0;
+
+ out_rtt:
+ arena->rtt[lane] = RTT_INVALID;
+ out_lane:
+ nd_region_release_lane(btt->nd_region, lane);
+ return ret;
+}
+
+static int btt_write_pg(struct btt *btt, sector_t sector, struct page *page,
+ unsigned int off, unsigned int len)
+{
+ int ret = 0;
+ struct arena_info *arena = NULL;
+ u32 premap = 0, old_postmap, new_postmap, lane = 0, i;
+ struct log_entry log;
+ int sub;
+
+ while (len) {
+ u32 cur_len;
+
+ lane = nd_region_acquire_lane(btt->nd_region);
+
+ ret = lba_to_arena(btt, sector, &premap, &arena);
+ if (ret)
+ goto out_lane;
+ cur_len = min(arena->external_lbasize, len);
+
+ if ((arena->flags & IB_FLAG_ERROR_MASK) != 0) {
+ ret = -EIO;
+ goto out_lane;
+ }
+
+ new_postmap = arena->freelist[lane].block;
+
+ /* Wait if the new block is being read from */
+ for (i = 0; i < arena->nfree; i++)
+ while (arena->rtt[i] == (RTT_VALID | new_postmap))
+ cpu_relax();
+
+
+ if (new_postmap >= arena->internal_nlba) {
+ ret = -EIO;
+ goto out_lane;
+ } else
+ ret = btt_data_write(arena, new_postmap, page,
+ off, cur_len);
+ if (ret)
+ goto out_lane;
+
+ lock_map(arena, premap);
+ ret = btt_map_read(arena, premap, &old_postmap, NULL, NULL);
+ if (ret)
+ goto out_map;
+ if (old_postmap >= arena->internal_nlba) {
+ ret = -EIO;
+ goto out_map;
+ }
+
+ log.lba = cpu_to_le32(premap);
+ log.old_map = cpu_to_le32(old_postmap);
+ log.new_map = cpu_to_le32(new_postmap);
+ log.seq = cpu_to_le32(arena->freelist[lane].seq);
+ sub = arena->freelist[lane].sub;
+ ret = btt_flog_write(arena, lane, sub, &log);
+ if (ret)
+ goto out_map;
+
+ ret = btt_map_write(arena, premap, new_postmap, 0, 0);
+ if (ret)
+ goto out_map;
+
+ unlock_map(arena, premap);
+ nd_region_release_lane(btt->nd_region, lane);
+
+ len -= cur_len;
+ off += cur_len;
+ sector += arena->external_lbasize >> SECTOR_SHIFT;
+ }
+
+ return 0;
+
+ out_map:
+ unlock_map(arena, premap);
+ out_lane:
+ nd_region_release_lane(btt->nd_region, lane);
+ return ret;
+}
+
+static int btt_do_bvec(struct btt *btt, struct page *page,
+ unsigned int len, unsigned int off, int rw,
+ sector_t sector)
+{
+ int ret;
+
+ if (rw == READ) {
+ ret = btt_read_pg(btt, page, off, sector, len);
+ flush_dcache_page(page);
+ } else {
+ flush_dcache_page(page);
+ ret = btt_write_pg(btt, sector, page, off, len);
+ }
+
+ return ret;
+}
+
+static void btt_make_request(struct request_queue *q, struct bio *bio)
+{
+ struct block_device *bdev = bio->bi_bdev;
+ struct btt *btt = q->queuedata;
+ int rw;
+ struct bio_vec bvec;
+ sector_t sector;
+ struct bvec_iter iter;
+ int err = 0;
+
+ sector = bio->bi_iter.bi_sector;
+ if (bio_end_sector(bio) > get_capacity(bdev->bd_disk)) {
+ err = -EIO;
+ goto out;
+ }
+
+ BUG_ON(bio->bi_rw & REQ_DISCARD);
+
+ rw = bio_rw(bio);
+ if (rw == READA)
+ rw = READ;
+
+ bio_for_each_segment(bvec, bio, iter) {
+ unsigned int len = bvec.bv_len;
+
+ BUG_ON(len > PAGE_SIZE);
+ /* Make sure len is in multiples of lbasize. */
+ /* XXX is this right? */
+ BUG_ON(len < btt->lbasize);
+ BUG_ON(len % btt->lbasize);
+
+ err = btt_do_bvec(btt, bvec.bv_page, len, bvec.bv_offset,
+ rw, sector);
+ if (err) {
+ dev_info(&btt->nd_btt->dev,
+ "io error in %s sector %lld, len %d,\n",
+ (rw == READ) ? "READ" : "WRITE",
+ (unsigned long long) sector, len);
+ goto out;
+ }
+ sector += len >> SECTOR_SHIFT;
+ }
+
+out:
+ bio_endio(bio, err);
+}
+
+static int btt_getgeo(struct block_device *bd, struct hd_geometry *geo)
+{
+ /* some standard values */
+ geo->heads = 1 << 6;
+ geo->sectors = 1 << 5;
+ geo->cylinders = get_capacity(bd->bd_disk) >> 11;
+ return 0;
+}
+
+static const struct block_device_operations btt_fops = {
+ .owner = THIS_MODULE,
+ /* TODO: Disable rw_page till lazy init is reworked */
+ /*.rw_page = btt_rw_page, */
+ .getgeo = btt_getgeo,
+};
+
+static int btt_blk_init(struct btt *btt)
+{
+ int ret;
+
+ /* create a new disk and request queue for btt */
+ btt->btt_queue = blk_alloc_queue(GFP_KERNEL);
+ if (!btt->btt_queue)
+ return -ENOMEM;
+
+ btt->btt_disk = alloc_disk(0);
+ if (!btt->btt_disk) {
+ ret = -ENOMEM;
+ goto out_free_queue;
+ }
+
+ sprintf(btt->btt_disk->disk_name, "nd%d", btt->nd_btt->id);
+ btt->btt_disk->driverfs_dev = &btt->nd_btt->dev;
+ btt->btt_disk->major = btt_major;
+ btt->btt_disk->first_minor = 0;
+ btt->btt_disk->fops = &btt_fops;
+ btt->btt_disk->private_data = btt;
+ btt->btt_disk->queue = btt->btt_queue;
+ btt->btt_disk->flags = GENHD_FL_EXT_DEVT;
+
+ blk_queue_make_request(btt->btt_queue, btt_make_request);
+ blk_queue_max_hw_sectors(btt->btt_queue, 1024);
+ blk_queue_bounce_limit(btt->btt_queue, BLK_BOUNCE_ANY);
+ blk_queue_logical_block_size(btt->btt_queue, btt->lbasize);
+ btt->btt_queue->queuedata = btt;
+
+ set_capacity(btt->btt_disk, btt->nlba * btt->lbasize >> SECTOR_SHIFT);
+ add_disk(btt->btt_disk);
+
+ return 0;
+
+out_free_queue:
+ blk_cleanup_queue(btt->btt_queue);
+ return ret;
+}
+
+static void btt_blk_cleanup(struct btt *btt)
+{
+ del_gendisk(btt->btt_disk);
+ put_disk(btt->btt_disk);
+ blk_cleanup_queue(btt->btt_queue);
+}
+
+/**
+ * btt_init - initialize a block translation table for the given device
+ * @nd_btt: device with BTT geometry and backing device info
+ * @rawsize: raw size in bytes of the backing device
+ * @lbasize: lba size of the backing device
+ * @uuid: A uuid for the backing device - this is stored on media
+ * @maxlane: maximum number of parallel requests the device can handle
+ *
+ * Initialize a Block Translation Table on a backing device to provide
+ * single sector power fail atomicity.
+ *
+ * Context:
+ * Might sleep.
+ *
+ * Returns:
+ * Pointer to a new struct btt on success, NULL on failure.
+ */
+static struct btt *btt_init(struct nd_btt *nd_btt, unsigned long long rawsize,
+ u32 lbasize, u8 *uuid, struct nd_region *nd_region)
+{
+ int ret;
+ struct btt *btt;
+ struct device *dev = &nd_btt->dev;
+
+ btt = kzalloc(sizeof(struct btt), GFP_KERNEL);
+ if (!btt)
+ return NULL;
+
+ btt->nd_btt = nd_btt;
+ btt->rawsize = rawsize;
+ btt->lbasize = lbasize;
+ INIT_LIST_HEAD(&btt->arena_list);
+ mutex_init(&btt->init_lock);
+ btt->nd_region = nd_region;
+
+ ret = discover_arenas(btt);
+ if (ret) {
+ dev_err(dev, "init: error in arena_discover: %d\n", ret);
+ goto out_free;
+ }
+
+ if (btt->init_state != INIT_READY) {
+ btt->num_arenas = (rawsize / ARENA_MAX_SIZE) +
+ ((rawsize % ARENA_MAX_SIZE) ? 1 : 0);
+ dev_dbg(dev, "init: %d arenas for %llu rawsize\n",
+ btt->num_arenas, rawsize);
+
+ ret = create_arenas(btt);
+ if (ret) {
+ dev_info(dev, "init: create_arenas: %d\n", ret);
+ goto out_free;
+ }
+
+ ret = btt_meta_init(btt);
+ if (ret) {
+ dev_err(dev, "init: error in meta_init: %d\n", ret);
+ return NULL;
+ }
+ }
+
+ ret = btt_blk_init(btt);
+ if (ret) {
+ dev_err(dev, "init: error in blk_init: %d\n", ret);
+ goto out_free;
+ }
+
+ btt_debugfs_init(btt);
+
+ return btt;
+
+ out_free:
+ kfree(btt);
+ return NULL;
+}
+
+/**
+ * btt_fini - de-initialize a BTT
+ * @btt: the BTT handle that was generated by btt_init
+ *
+ * De-initialize a Block Translation Table on device removal
+ *
+ * Context:
+ * Might sleep.
+ */
+static void btt_fini(struct btt *btt)
+{
+ if (btt) {
+ btt_blk_cleanup(btt);
+ free_arenas(btt);
+ debugfs_remove_recursive(btt->debugfs_dir);
+ kfree(btt);
+ }
+}
+
+static int link_btt(struct nd_btt *nd_btt)
+{
+ struct block_device *bdev = nd_btt->backing_dev;
+ struct kobject *dir = &part_to_dev(bdev->bd_part)->kobj;
+
+ return sysfs_create_link(dir, &nd_btt->dev.kobj, "nd_btt");
+}
+
+static void unlink_btt(struct nd_btt *nd_btt)
+{
+ struct block_device *bdev = nd_btt->backing_dev;
+ struct kobject *dir;
+
+ /* if backing_dev was deleted first we may have nothing to unlink */
+ if (!nd_btt->backing_dev)
+ return;
+
+ dir = &part_to_dev(bdev->bd_part)->kobj;
+ sysfs_remove_link(dir, "nd_btt");
+}
+
+static int nd_btt_probe(struct device *dev)
+{
+ struct nd_btt *nd_btt = to_nd_btt(dev);
+ struct nd_io_claim *ndio_claim = nd_btt->ndio_claim;
+ struct nd_region *nd_region;
+ struct block_device *bdev;
+ struct btt *btt;
+ size_t rawsize;
+ int rc;
+
+ if (!ndio_claim || !nd_btt->uuid || !nd_btt->backing_dev
+ || !nd_btt->lbasize)
+ return -ENODEV;
+
+ rc = link_btt(nd_btt);
+ if (rc)
+ return rc;
+
+ bdev = nd_btt->backing_dev;
+ /* the first 4K of a device is padding */
+ nd_btt->offset = nd_partition_offset(bdev) + SZ_4K;
+ rawsize = (bdev->bd_part->nr_sects << SECTOR_SHIFT) - SZ_4K;
+ if (rawsize < ARENA_MIN_SIZE) {
+ rc = -ENXIO;
+ goto err_btt;
+ }
+ nd_btt->ndio = nd_btt->ndio_claim->parent;
+ nd_region = to_nd_region(nd_btt->ndio->dev->parent);
+ btt = btt_init(nd_btt, rawsize, nd_btt->lbasize, nd_btt->uuid,
+ nd_region);
+ if (!btt) {
+ rc = -ENOMEM;
+ goto err_btt;
+ }
+ btt->backing_dev = bdev;
+ dev_set_drvdata(dev, btt);
+
+ return 0;
+ err_btt:
+ unlink_btt(nd_btt);
+ return rc;
+}
+
+static int nd_btt_remove(struct device *dev)
+{
+ struct nd_btt *nd_btt = to_nd_btt(dev);
+ struct btt *btt = dev_get_drvdata(dev);
+
+ btt_fini(btt);
+ unlink_btt(nd_btt);
+
+ return 0;
+}
+
+static struct nd_device_driver nd_btt_driver = {
+ .probe = nd_btt_probe,
+ .remove = nd_btt_remove,
+ .drv = {
+ .name = "nd_btt",
+ },
+ .type = ND_DRIVER_BTT,
+};
+
+static int __init nd_btt_init(void)
+{
+ int rc;
+
+ BUILD_BUG_ON(sizeof(struct btt_sb) != SZ_4K);
+
+ btt_major = register_blkdev(0, "btt");
+ if (btt_major < 0)
+ return btt_major;
+
+ debugfs_root = debugfs_create_dir("btt", NULL);
+ if (IS_ERR_OR_NULL(debugfs_root)) {
+ rc = -ENXIO;
+ goto err_debugfs;
+ }
+
+ rc = nd_driver_register(&nd_btt_driver);
+ if (rc < 0)
+ goto err_driver;
+ return 0;
+
+ err_driver:
+ debugfs_remove_recursive(debugfs_root);
+ err_debugfs:
+ unregister_blkdev(btt_major, "btt");
+
+ return rc;
+}
+
+static void __exit nd_btt_exit(void)
+{
+ driver_unregister(&nd_btt_driver.drv);
+ debugfs_remove_recursive(debugfs_root);
+ unregister_blkdev(btt_major, "btt");
+}
+
+MODULE_ALIAS_ND_DEVICE(ND_DEVICE_BTT);
+MODULE_AUTHOR("Vishal Verma <[email protected]>");
+MODULE_LICENSE("GPL v2");
+module_init(nd_btt_init);
+module_exit(nd_btt_exit);
diff --git a/drivers/block/nd/btt.h b/drivers/block/nd/btt.h
index e8f6d8e0ddd3..d4e67c75c91f 100644
--- a/drivers/block/nd/btt.h
+++ b/drivers/block/nd/btt.h
@@ -19,6 +19,39 @@
#define BTT_SIG_LEN 16
#define BTT_SIG "BTT_ARENA_INFO\0"
+#define MAP_ENT_SIZE 4
+#define MAP_TRIM_SHIFT 31
+#define MAP_TRIM_MASK (1 << MAP_TRIM_SHIFT)
+#define MAP_ERR_SHIFT 30
+#define MAP_ERR_MASK (1 << MAP_ERR_SHIFT)
+#define MAP_LBA_MASK (~((1 << MAP_TRIM_SHIFT) | (1 << MAP_ERR_SHIFT)))
+#define MAP_ENT_NORMAL 0xC0000000
+#define LOG_ENT_SIZE sizeof(struct log_entry)
+#define ARENA_MIN_SIZE (1UL << 24) /* 16 MB */
+#define ARENA_MAX_SIZE (1ULL << 39) /* 512 GB */
+#define RTT_VALID (1UL << 31)
+#define RTT_INVALID 0
+#define INT_LBASIZE_ALIGNMENT 256
+#define BTT_PG_SIZE 4096
+#define BTT_DEFAULT_NFREE ND_MAX_LANES
+#define LOG_SEQ_INIT 1
+
+#define IB_FLAG_ERROR 0x00000001
+#define IB_FLAG_ERROR_MASK 0x00000001
+
+enum btt_init_state {
+ INIT_UNCHECKED = 0,
+ INIT_NOTFOUND,
+ INIT_READY
+};
+
+struct log_entry {
+ __le32 lba;
+ __le32 old_map;
+ __le32 new_map;
+ __le32 seq;
+ __le64 padding[2];
+};
struct btt_sb {
u8 signature[BTT_SIG_LEN];
@@ -42,4 +75,111 @@ struct btt_sb {
__le64 checksum;
};
+struct free_entry {
+ u32 block;
+ u8 sub;
+ u8 seq;
+};
+
+struct aligned_lock {
+ union {
+ spinlock_t lock;
+ u8 cacheline_padding[L1_CACHE_BYTES];
+ };
+};
+
+/**
+ * struct arena_info - handle for an arena
+ * @size: Size in bytes this arena occupies on the raw device.
+ * This includes arena metadata.
+ * @external_lba_start: The first external LBA in this arena.
+ * @internal_nlba: Number of internal blocks available in the arena
+ * including nfree reserved blocks
+ * @internal_lbasize: Internal and external lba sizes may be different as
+ * we can round up 'odd' external lbasizes such as 520B
+ * to be aligned.
+ * @external_nlba: Number of blocks contributed by the arena to the number
+ * reported to upper layers. (internal_nlba - nfree)
+ * @external_lbasize: LBA size as exposed to upper layers.
+ * @nfree: A reserve number of 'free' blocks that is used to
+ * handle incoming writes.
+ * @version_major: Metadata layout version major.
+ * @version_minor: Metadata layout version minor.
+ * @nextoff: Offset in bytes to the start of the next arena.
+ * @infooff: Offset in bytes to the info block of this arena.
+ * @dataoff: Offset in bytes to the data area of this arena.
+ * @mapoff: Offset in bytes to the map area of this arena.
+ * @logoff: Offset in bytes to the log area of this arena.
+ * @info2off: Offset in bytes to the backup info block of this arena.
+ * @freelist: Pointer to in-memory list of free blocks
+ * @rtt: Pointer to in-memory "Read Tracking Table"
+ * @map_locks: Spinlocks protecting concurrent map writes
+ * @nd_btt: Pointer to parent nd_btt structure.
+ * @list: List head for list of arenas
+ * @debugfs_dir: Debugfs dentry
+ * @flags: Arena flags - may signify error states.
+ *
+ * arena_info is a per-arena handle. Once an arena is narrowed down for an
+ * IO, this struct is passed around for the duration of the IO.
+ */
+struct arena_info {
+ u64 size; /* Total bytes for this arena */
+ u64 external_lba_start;
+ u32 internal_nlba;
+ u32 internal_lbasize;
+ u32 external_nlba;
+ u32 external_lbasize;
+ u32 nfree;
+ u16 version_major;
+ u16 version_minor;
+ /* Byte offsets to the different on-media structures */
+ u64 nextoff;
+ u64 infooff;
+ u64 dataoff;
+ u64 mapoff;
+ u64 logoff;
+ u64 info2off;
+ /* Pointers to other in-memory structures for this arena */
+ struct free_entry *freelist;
+ u32 *rtt;
+ struct aligned_lock *map_locks;
+ struct nd_btt *nd_btt;
+ struct list_head list;
+ struct dentry *debugfs_dir;
+ /* Arena flags */
+ u32 flags;
+};
+
+/**
+ * struct btt - handle for a BTT instance
+ * @btt_disk: Pointer to the gendisk for BTT device
+ * @btt_queue: Pointer to the request queue for the BTT device
+ * @arena_list: Head of the list of arenas
+ * @debugfs_dir: Debugfs dentry
+ * @backing_dev: Backing block device for the BTT
+ * @nd_btt: Parent nd_btt struct
+ * @nlba: Number of logical blocks exposed to the upper layers
+ * after removing the amount of space needed by metadata
+ * @rawsize: Total size in bytes of the available backing device
+ * @lbasize: LBA size as requested and presented to upper layers
+ * @lanes: Per-lane spinlocks
+ * @init_lock: Mutex used for the BTT initialization
+ * @init_state: Flag describing the initialization state for the BTT
+ * @num_arenas: Number of arenas in the BTT instance
+ */
+struct btt {
+ struct gendisk *btt_disk;
+ struct request_queue *btt_queue;
+ struct list_head arena_list;
+ struct dentry *debugfs_dir;
+ struct block_device *backing_dev;
+ struct nd_btt *nd_btt;
+ u64 nlba;
+ unsigned long long rawsize;
+ u32 lbasize;
+ struct nd_region *nd_region;
+ struct mutex init_lock;
+ int init_state;
+ int num_arenas;
+};
#endif
diff --git a/drivers/block/nd/btt_devs.c b/drivers/block/nd/btt_devs.c
index e6f0b8b999d8..0746db70973c 100644
--- a/drivers/block/nd/btt_devs.c
+++ b/drivers/block/nd/btt_devs.c
@@ -342,7 +342,8 @@ struct nd_btt *nd_btt_create(struct nd_bus *nd_bus)
*/
u64 nd_btt_sb_checksum(struct btt_sb *btt_sb)
{
- u64 sum, sum_save;
+ u64 sum;
+ __le64 sum_save;
sum_save = btt_sb->checksum;
btt_sb->checksum = 0;
diff --git a/drivers/block/nd/libnd.h b/drivers/block/nd/libnd.h
index 3f6b5e09cd67..e188840ed2b9 100644
--- a/drivers/block/nd/libnd.h
+++ b/drivers/block/nd/libnd.h
@@ -76,6 +76,7 @@ struct nd_region_desc {
const struct attribute_group **attr_groups;
struct nd_interleave_set *nd_set;
void *provider_data;
+ int num_lanes;
};
struct nd_bus;
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
index 4ab4765dc9ee..68e9ec824dc8 100644
--- a/drivers/block/nd/nd-private.h
+++ b/drivers/block/nd/nd-private.h
@@ -76,6 +76,7 @@ int __init nd_bus_init(void);
void nd_bus_exit(void);
int __init nd_dimm_init(void);
int __init nd_region_init(void);
+void __init nd_region_init_locks(void);
void nd_dimm_exit(void);
int nd_region_exit(void);
void nd_region_probe_start(struct nd_bus *nd_bus, struct device *dev);
diff --git a/drivers/block/nd/nd.h b/drivers/block/nd/nd.h
index c6ed26d4dcad..a29fb7409925 100644
--- a/drivers/block/nd/nd.h
+++ b/drivers/block/nd/nd.h
@@ -22,6 +22,12 @@
#include "label.h"
enum {
+ /*
+ * Limits the maximum number of block apertures a dimm can
+ * support and is an input to the geometry/on-disk-format of a
+ * BTT instance
+ */
+ ND_MAX_LANES = 256,
SECTOR_SHIFT = 9,
};
@@ -101,7 +107,7 @@ struct nd_region {
u16 ndr_mappings;
u64 ndr_size;
u64 ndr_start;
- int id;
+ int id, num_lanes;
void *provider_data;
struct nd_interleave_set *nd_set;
struct nd_mapping mapping[0];
@@ -226,6 +232,8 @@ struct nd_btt *to_nd_btt(struct device *dev);
struct btt_sb;
u64 nd_btt_sb_checksum(struct btt_sb *btt_sb);
struct nd_region *to_nd_region(struct device *dev);
+unsigned int nd_region_acquire_lane(struct nd_region *nd_region);
+void nd_region_release_lane(struct nd_region *nd_region, unsigned int lane);
int nd_region_to_namespace_type(struct nd_region *nd_region);
int nd_region_register_namespaces(struct nd_region *nd_region, int *err);
u64 nd_region_interleave_set_cookie(struct nd_region *nd_region);
diff --git a/drivers/block/nd/region.c b/drivers/block/nd/region.c
index 9d1fd45d78a1..dd5a885cea11 100644
--- a/drivers/block/nd/region.c
+++ b/drivers/block/nd/region.c
@@ -15,6 +15,72 @@
#include <linux/nd.h>
#include "nd.h"
+static struct {
+ struct {
+ int count[CONFIG_ND_MAX_REGIONS];
+ spinlock_t lock[CONFIG_ND_MAX_REGIONS];
+ } lane[NR_CPUS];
+} nd_percpu_lane;
+
+static void __init nd_region_init_locks(void)
+{
+ int i, j;
+
+ for (i = 0; i < NR_CPUS; i++)
+ for (j = 0; j < CONFIG_ND_MAX_REGIONS; j++)
+ spin_lock_init(&nd_percpu_lane.lane[i].lock[j]);
+}
+
+/**
+ * nd_region_acquire_lane - allocate and lock a lane
+ * @nd_region: region id and number of lanes possible
+ *
+ * A lane correlates to a BLK-data-window and/or a log slot in the BTT.
+ * We optimize for the common case where there are 256 lanes, one
+ * per-cpu. For larger systems we need to lock to share lanes. For now
+ * this implementation assumes the cost of maintaining an allocator for
+ * free lanes is on the order of the lock hold time, so it implements a
+ * static lane = cpu % num_lanes mapping.
+ *
+ * In the case of a BTT instance on top of a BLK namespace a lane may be
+ * acquired recursively. We lock on the first instance.
+ *
+ * In the case of a BTT instance on top of PMEM, we only acquire a lane
+ * for the BTT metadata updates.
+ */
+unsigned int nd_region_acquire_lane(struct nd_region *nd_region)
+{
+ unsigned int cpu, lane;
+
+ cpu = get_cpu();
+
+ if (nd_region->num_lanes < NR_CPUS) {
+ unsigned int id = nd_region->id;
+
+ lane = cpu % nd_region->num_lanes;
+ if (nd_percpu_lane.lane[cpu].count[id]++ == 0)
+ spin_lock(&nd_percpu_lane.lane[lane].lock[id]);
+ } else
+ lane = cpu;
+
+ return lane;
+}
+EXPORT_SYMBOL(nd_region_acquire_lane);
+
+void nd_region_release_lane(struct nd_region *nd_region, unsigned int lane)
+{
+ if (nd_region->num_lanes < NR_CPUS) {
+ unsigned int cpu = get_cpu();
+ unsigned int id = nd_region->id;
+
+ if (--nd_percpu_lane.lane[cpu].count[id] == 0)
+ spin_unlock(&nd_percpu_lane.lane[lane].lock[id]);
+ put_cpu();
+ }
+ put_cpu();
+}
+EXPORT_SYMBOL(nd_region_release_lane);
+
static int nd_region_probe(struct device *dev)
{
int err;
@@ -76,6 +142,7 @@ static struct nd_device_driver nd_region_driver = {
int __init nd_region_init(void)
{
+ nd_region_init_locks();
return nd_driver_register(&nd_region_driver);
}
diff --git a/drivers/block/nd/region_devs.c b/drivers/block/nd/region_devs.c
index bcdd8e1e21a2..268d9ef67f9c 100644
--- a/drivers/block/nd/region_devs.c
+++ b/drivers/block/nd/region_devs.c
@@ -538,6 +538,12 @@ static noinline struct nd_region *nd_region_create(struct nd_bus *nd_bus,
if (nd_region->id < 0) {
kfree(nd_region);
return NULL;
+ } else if (nd_region->id >= CONFIG_ND_MAX_REGIONS) {
+ dev_err(&nd_bus->dev, "max region limit %d reached\n",
+ CONFIG_ND_MAX_REGIONS);
+ ida_simple_remove(®ion_ida, nd_region->id);
+ kfree(nd_region);
+ return NULL;
}
memcpy(nd_region->mapping, ndr_desc->nd_mapping,
@@ -551,6 +557,7 @@ static noinline struct nd_region *nd_region_create(struct nd_bus *nd_bus,
nd_region->ndr_mappings = ndr_desc->num_mappings;
nd_region->provider_data = ndr_desc->provider_data;
nd_region->nd_set = ndr_desc->nd_set;
+ nd_region->num_lanes = ndr_desc->num_lanes;
ida_init(&nd_region->ns_ida);
dev = &nd_region->dev;
dev_set_name(dev, "region%d", nd_region->id);
@@ -567,6 +574,7 @@ static noinline struct nd_region *nd_region_create(struct nd_bus *nd_bus,
struct nd_region *nd_pmem_region_create(struct nd_bus *nd_bus,
struct nd_region_desc *ndr_desc)
{
+ ndr_desc->num_lanes = ND_MAX_LANES;
return nd_region_create(nd_bus, ndr_desc, &nd_pmem_device_type);
}
EXPORT_SYMBOL_GPL(nd_pmem_region_create);
@@ -576,6 +584,7 @@ struct nd_region *nd_blk_region_create(struct nd_bus *nd_bus,
{
if (ndr_desc->num_mappings > 1)
return NULL;
+ ndr_desc->num_lanes = min(ndr_desc->num_lanes, ND_MAX_LANES);
return nd_region_create(nd_bus, ndr_desc, &nd_blk_device_type);
}
EXPORT_SYMBOL_GPL(nd_blk_region_create);
@@ -583,6 +592,7 @@ EXPORT_SYMBOL_GPL(nd_blk_region_create);
struct nd_region *nd_volatile_region_create(struct nd_bus *nd_bus,
struct nd_region_desc *ndr_desc)
{
+ ndr_desc->num_lanes = ND_MAX_LANES;
return nd_region_create(nd_bus, ndr_desc, &nd_volatile_device_type);
}
EXPORT_SYMBOL_GPL(nd_volatile_region_create);
From: Ross Zwisler <[email protected]>
The libnd implementation handles allocating dimm address space (DPA)
between PMEM and BLK mode interfaces. After DPA has been allocated from
a BLK-region to a BLK-namespace the nd_blk driver attaches to handle I/O
as a struct bio based block device. Unlike PMEM, BLK is required to
handle platform specific details like mmio register formats and memory
controller interleave. For this reason the libnd generic nd_blk driver
calls back into the bus provider to carry out the I/O.
This initial implementation handles the BLK interface defined by the
ACPI 6 NFIT [1] and the NVDIMM DSM Interface Example [2] composed from
DCR (dimm control region), BDW (block data window), IDT (interleave
descriptor) NFIT structures and the hardware register format.
[1]: http://www.uefi.org/sites/default/files/resources/ACPI_6.0.pdf
[2]: http://pmem.io/documents/NVDIMM_DSM_Interface_Example.pdf
Cc: Andy Lutomirski <[email protected]>
Cc: Boaz Harrosh <[email protected]>
Cc: H. Peter Anvin <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Signed-off-by: Ross Zwisler <[email protected]>
Signed-off-by: Dan Williams <[email protected]>
---
drivers/block/nd/Kconfig | 12 +
drivers/block/nd/Makefile | 3
drivers/block/nd/acpi.c | 422 +++++++++++++++++++++++++++++++++++--
drivers/block/nd/acpi_nfit.h | 47 ++++
drivers/block/nd/blk.c | 264 +++++++++++++++++++++++
drivers/block/nd/libnd.h | 11 +
drivers/block/nd/namespace_devs.c | 47 ++++
drivers/block/nd/nd-private.h | 3
drivers/block/nd/nd.h | 16 +
drivers/block/nd/region.c | 8 +
drivers/block/nd/region_devs.c | 65 +++++-
drivers/block/nd/test/nfit.c | 29 +++
drivers/block/nd/test/nfit_test.h | 2
13 files changed, 891 insertions(+), 38 deletions(-)
create mode 100644 drivers/block/nd/blk.c
diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
index 612bf2b14283..bac4290129fc 100644
--- a/drivers/block/nd/Kconfig
+++ b/drivers/block/nd/Kconfig
@@ -95,6 +95,18 @@ config BLK_DEV_PMEM
Say Y if you want to use a NVDIMM described by ACPI, E820, etc...
+config ND_BLK
+ tristate "BLK: Block data window (aperture) device support"
+ depends on LIBND
+ default ND_ACPI
+ help
+ This driver performs I/O using a set of mmio windows on a
+ dimm. The set of apertures will all access the one DIMM.
+ Multiple windows allow multiple threads to have a different
+ portions of the dimm open at one time.
+
+ Say Y if you want to use a NVDIMM with BLK-mode capability
+
config ND_BTT_DEVS
bool
diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
index 7d778b4523d4..ef36927618e5 100644
--- a/drivers/block/nd/Makefile
+++ b/drivers/block/nd/Makefile
@@ -18,6 +18,7 @@ obj-$(CONFIG_ND_E820) += nd_e820.o
obj-$(CONFIG_NFIT_TEST) += test/
obj-$(CONFIG_BLK_DEV_PMEM) += nd_pmem.o
obj-$(CONFIG_ND_BTT) += nd_btt.o
+obj-$(CONFIG_ND_BLK) += nd_blk.o
nd_acpi-y := acpi.o
@@ -27,6 +28,8 @@ nd_pmem-y := pmem.o
nd_btt-y := btt.o
+nd_blk-y := blk.o
+
libnd-y := core.o
libnd-y += bus.o
libnd-y += dimm_devs.o
diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
index 5b9997fbc344..e4ff3a9b4fc1 100644
--- a/drivers/block/nd/acpi.c
+++ b/drivers/block/nd/acpi.c
@@ -12,12 +12,14 @@
*/
#include <linux/list_sort.h>
#include <linux/module.h>
+#include <linux/mutex.h>
#include <linux/ndctl.h>
#include <linux/list.h>
#include <linux/acpi.h>
#include <linux/sort.h>
#include "acpi_nfit.h"
#include "libnd.h"
+#include "nd.h"
static bool warn_checksum;
module_param(warn_checksum, bool, S_IRUGO|S_IWUSR);
@@ -84,7 +86,7 @@ static int nd_acpi_ctl(struct nd_bus_descriptor *nd_desc,
if (!adev)
return -ENOTTY;
- dimm_name = dev_name(&adev->dev);
+ dimm_name = nd_dimm_name(nd_dimm);
cmd_name = nd_dimm_cmd_name(cmd);
dsm_mask = nfit_mem->dsm_mask;
desc = nd_cmd_dimm_desc(cmd);
@@ -301,10 +303,21 @@ static void *add_table(struct acpi_nfit_desc *acpi_desc, void *table, const void
bdw->dcr_index, bdw->num_bdw);
break;
}
- /* TODO */
- case NFIT_TABLE_IDT:
- dev_dbg(dev, "%s: idt\n", __func__);
+ case NFIT_TABLE_IDT: {
+ struct nfit_idt *nfit_idt = devm_kzalloc(dev, sizeof(*nfit_idt),
+ GFP_KERNEL);
+ struct acpi_nfit_idt *idt = table;
+
+ if (!nfit_idt)
+ return err;
+ INIT_LIST_HEAD(&nfit_idt->list);
+ nfit_idt->idt = idt;
+ list_add_tail(&nfit_idt->list, &acpi_desc->idts);
+ dev_dbg(dev, "%s: idt index: %d num_lines: %d\n", __func__,
+ idt->idt_index, idt->num_lines);
break;
+ }
+ /* TODO */
case NFIT_TABLE_FLUSH:
dev_dbg(dev, "%s: flush\n", __func__);
break;
@@ -356,8 +369,11 @@ static int nfit_mem_add(struct acpi_nfit_desc *acpi_desc,
struct nfit_mem *nfit_mem, struct acpi_nfit_spa *spa)
{
u16 dcr_index = __to_nfit_memdev(nfit_mem)->dcr_index;
+ struct nfit_memdev *nfit_memdev;
struct nfit_dcr *nfit_dcr;
struct nfit_bdw *nfit_bdw;
+ struct nfit_idt *nfit_idt;
+ u16 idt_index, spa_index;
list_for_each_entry(nfit_dcr, &acpi_desc->dcrs, list) {
if (nfit_dcr->dcr->dcr_index != dcr_index)
@@ -390,6 +406,26 @@ static int nfit_mem_add(struct acpi_nfit_desc *acpi_desc,
return 0;
nfit_mem_find_spa_bdw(acpi_desc, nfit_mem);
+
+ if (!nfit_mem->spa_bdw)
+ return 0;
+
+ spa_index = nfit_mem->spa_bdw->spa_index;
+ list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) {
+ if (nfit_memdev->memdev->spa_index != spa_index ||
+ nfit_memdev->memdev->dcr_index != dcr_index)
+ continue;
+ nfit_mem->memdev_bdw = nfit_memdev->memdev;
+ idt_index = nfit_memdev->memdev->idt_index;
+ list_for_each_entry(nfit_idt, &acpi_desc->idts, list) {
+ if (nfit_idt->idt->idt_index != idt_index)
+ continue;
+ nfit_mem->idt_bdw = nfit_idt->idt;
+ break;
+ }
+ break;
+ }
+
return 0;
}
@@ -433,9 +469,19 @@ static int nfit_mem_dcr_init(struct acpi_nfit_desc *acpi_desc,
}
if (type == NFIT_SPA_DCR) {
+ struct nfit_idt *nfit_idt;
+ u16 idt_index;
+
/* multiple dimms may share a SPA when interleaved */
nfit_mem->spa_dcr = spa;
nfit_mem->memdev_dcr = nfit_memdev->memdev;
+ idt_index = nfit_memdev->memdev->idt_index;
+ list_for_each_entry(nfit_idt, &acpi_desc->idts, list) {
+ if (nfit_idt->idt->idt_index != idt_index)
+ continue;
+ nfit_mem->idt_dcr = nfit_idt->idt;
+ break;
+ }
} else {
/*
* A single dimm may belong to multiple SPA-PM
@@ -756,7 +802,7 @@ static ssize_t spa_index_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
struct nd_region *nd_region = to_nd_region(dev);
- struct nfit_spa *nfit_spa = nd_region_provider_data(nd_region);
+ struct nfit_spa *nfit_spa = nd_region->provider_data;
return sprintf(buf, "%d\n", nfit_spa->spa->spa_index);
}
@@ -864,13 +910,343 @@ static int nd_acpi_init_interleave_set(struct acpi_nfit_desc *acpi_desc,
return 0;
}
+static u64 to_interleave_offset(u64 offset, struct nfit_blk_mmio *mmio)
+{
+ struct acpi_nfit_idt *idt = mmio->idt;
+ u32 sub_line_offset, line_index, line_offset;
+ u64 line_no, table_skip_count, table_offset;
+
+ line_no = div_u64_rem(offset, mmio->line_size, &sub_line_offset);
+ table_skip_count = div_u64_rem(line_no, mmio->num_lines, &line_index);
+ line_offset = idt->line_offset[line_index]
+ * mmio->line_size;
+ table_offset = table_skip_count * mmio->table_size;
+
+ return mmio->base_offset + line_offset + table_offset + sub_line_offset;
+}
+
+static u64 read_blk_stat(struct nfit_blk *nfit_blk, unsigned int bw)
+{
+ struct nfit_blk_mmio *mmio = &nfit_blk->mmio[DCR];
+ u64 offset = nfit_blk->stat_offset + mmio->size * bw;
+
+ if (mmio->num_lines)
+ offset = to_interleave_offset(offset, mmio);
+
+ return readq(mmio->base + offset);
+}
+
+static void write_blk_ctl(struct nfit_blk *nfit_blk, unsigned int bw,
+ resource_size_t dpa, unsigned int len, unsigned int write)
+{
+ u64 cmd, offset;
+ struct nfit_blk_mmio *mmio = &nfit_blk->mmio[DCR];
+
+ enum {
+ BCW_OFFSET_MASK = (1ULL << 48)-1,
+ BCW_LEN_SHIFT = 48,
+ BCW_LEN_MASK = (1ULL << 8) - 1,
+ BCW_CMD_SHIFT = 56,
+ };
+
+ cmd = (dpa >> L1_CACHE_SHIFT) & BCW_OFFSET_MASK;
+ len = len >> L1_CACHE_SHIFT;
+ cmd |= ((u64) len & BCW_LEN_MASK) << BCW_LEN_SHIFT;
+ cmd |= ((u64) write) << BCW_CMD_SHIFT;
+
+ offset = nfit_blk->cmd_offset + mmio->size * bw;
+ if (mmio->num_lines)
+ offset = to_interleave_offset(offset, mmio);
+
+ writeq(cmd, mmio->base + offset);
+ /* FIXME: conditionally perform read-back if mandated by firmware */
+}
+
+/* len is <= PAGE_SIZE by this point, so it can be done in a single BW I/O */
+static int nd_acpi_blk_region_do_io(struct nd_blk_region *ndbr, void *iobuf,
+ unsigned int len, int write, resource_size_t dpa)
+{
+ struct nfit_blk *nfit_blk = ndbr->blk_provider_data;
+ struct nfit_blk_mmio *mmio = &nfit_blk->mmio[BDW];
+ struct nd_region *nd_region = &ndbr->nd_region;
+ unsigned int bw, copied = 0;
+ u64 base_offset;
+ int rc;
+
+ bw = nd_region_acquire_lane(nd_region);
+ base_offset = nfit_blk->bdw_offset + dpa % L1_CACHE_BYTES + bw * mmio->size;
+ /* TODO: non-temporal access, flush hints, cache management etc... */
+ write_blk_ctl(nfit_blk, bw, dpa, len, write);
+ while (len) {
+ unsigned int c;
+ u64 offset;
+
+ if (mmio->num_lines) {
+ u32 line_offset;
+
+ offset = to_interleave_offset(base_offset + copied,
+ mmio);
+ div_u64_rem(offset, mmio->line_size, &line_offset);
+ c = min(len, mmio->line_size - line_offset);
+ } else {
+ offset = base_offset + nfit_blk->bdw_offset;
+ c = len;
+ }
+
+ if (write)
+ memcpy(mmio->base + offset, iobuf + copied, c);
+ else
+ memcpy(iobuf + copied, mmio->base + offset, c);
+
+ len -= c;
+ copied += c;
+ }
+ rc = read_blk_stat(nfit_blk, bw) ? -EIO : 0;
+ nd_region_release_lane(nd_region, bw);
+
+ return rc;
+}
+
+static void nfit_spa_mapping_release(struct kref *kref)
+{
+ struct nfit_spa_mapping *spa_map = to_spa_map(kref);
+ struct acpi_nfit_spa *spa = spa_map->spa;
+ struct acpi_nfit_desc *acpi_desc = spa_map->acpi_desc;
+
+ WARN_ON(!mutex_is_locked(&acpi_desc->spa_map_mutex));
+ dev_dbg(acpi_desc->dev, "%s: SPA%d\n", __func__, spa->spa_index);
+ iounmap(spa_map->iomem);
+ release_mem_region(spa->spa_base, spa->spa_length);
+ list_del(&spa_map->list);
+ kfree(spa_map);
+}
+
+static struct nfit_spa_mapping *find_spa_mapping(struct acpi_nfit_desc *acpi_desc,
+ struct acpi_nfit_spa *spa)
+{
+ struct nfit_spa_mapping *spa_map;
+
+ WARN_ON(!mutex_is_locked(&acpi_desc->spa_map_mutex));
+ list_for_each_entry(spa_map, &acpi_desc->spa_maps, list)
+ if (spa_map->spa == spa)
+ return spa_map;
+
+ return NULL;
+}
+
+static void nfit_spa_unmap(struct acpi_nfit_desc *acpi_desc,
+ struct acpi_nfit_spa *spa)
+{
+ struct nfit_spa_mapping *spa_map;
+
+ mutex_lock(&acpi_desc->spa_map_mutex);
+ spa_map = find_spa_mapping(acpi_desc, spa);
+
+ if (spa_map)
+ kref_put(&spa_map->kref, nfit_spa_mapping_release);
+ mutex_unlock(&acpi_desc->spa_map_mutex);
+}
+
+static void *__nfit_spa_map(struct acpi_nfit_desc *acpi_desc,
+ struct acpi_nfit_spa *spa)
+{
+ resource_size_t start = spa->spa_base;
+ resource_size_t n = spa->spa_length;
+ struct nfit_spa_mapping *spa_map;
+ struct resource *res;
+
+ WARN_ON(!mutex_is_locked(&acpi_desc->spa_map_mutex));
+
+ spa_map = find_spa_mapping(acpi_desc, spa);
+ if (spa_map) {
+ kref_get(&spa_map->kref);
+ return spa_map->iomem;
+ }
+
+ spa_map = kzalloc(sizeof(*spa_map), GFP_KERNEL);
+ if (!spa_map)
+ return NULL;
+
+ INIT_LIST_HEAD(&spa_map->list);
+ spa_map->spa = spa;
+ kref_init(&spa_map->kref);
+ spa_map->acpi_desc = acpi_desc;
+
+ res = request_mem_region(start, n, dev_name(acpi_desc->dev));
+ if (!res)
+ goto err_mem;
+
+ /* TODO: cacheability based on the spa type */
+ spa_map->iomem = ioremap_nocache(start, n);
+ if (!spa_map->iomem)
+ goto err_map;
+
+ list_add_tail(&spa_map->list, &acpi_desc->spa_maps);
+ return spa_map->iomem;
+
+ err_map:
+ release_mem_region(start, n);
+ err_mem:
+ kfree(spa_map);
+ return NULL;
+}
+
+/**
+ * nfit_spa_map - interleave-aware managed-mappings of acpi_nfit_spa ranges
+ * @nd_bus: NFIT-bus that provided the spa table entry
+ * @nfit_spa: spa table to map
+ *
+ * In the case where block-data-window apertures and
+ * dimm-control-regions are interleaved they will end up sharing a
+ * single request_mem_region() + ioremap() for the address range. In
+ * the style of devm nfit_spa_map() mappings are automatically dropped
+ * when all region devices referencing the same mapping are disabled /
+ * unbound.
+ */
+static void *nfit_spa_map(struct acpi_nfit_desc *acpi_desc,
+ struct acpi_nfit_spa *spa)
+{
+ struct nfit_spa_mapping *spa_map;
+
+ mutex_lock(&acpi_desc->spa_map_mutex);
+ spa_map = __nfit_spa_map(acpi_desc, spa);
+ mutex_unlock(&acpi_desc->spa_map_mutex);
+
+ return spa_map;
+}
+
+static int nfit_blk_init_interleave(struct nfit_blk_mmio *mmio,
+ struct acpi_nfit_idt *idt, u16 interleave_ways)
+{
+ if (idt) {
+ mmio->num_lines = idt->num_lines;
+ mmio->line_size = idt->line_size;
+ if (interleave_ways == 0)
+ return -ENXIO;
+ mmio->table_size = mmio->num_lines * interleave_ways
+ * mmio->line_size;
+ }
+
+ return 0;
+}
+
+int nd_acpi_blk_region_enable(struct nd_bus *nd_bus, struct nd_blk_region *ndbr)
+{
+ struct nd_bus_descriptor *nd_desc = to_nd_desc(nd_bus);
+ struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc);
+ struct nd_region *nd_region = &ndbr->nd_region;
+ struct nd_mapping *nd_mapping;
+ struct nfit_blk_mmio *mmio;
+ struct nfit_blk *nfit_blk;
+ struct nfit_mem *nfit_mem;
+ struct nd_dimm *nd_dimm;
+ int rc;
+
+ nd_mapping = &nd_region->mapping[0];
+ nd_dimm = nd_mapping->nd_dimm;
+ nfit_mem = nd_dimm_provider_data(nd_dimm);
+ if (!nfit_mem || !nfit_mem->dcr || !nfit_mem->bdw) {
+ dev_dbg(&nd_region->dev, "%s: missing%s%s%s\n", __func__,
+ nfit_mem ? "" : " nfit_mem",
+ nfit_mem->dcr ? "" : " dcr",
+ nfit_mem->bdw ? "" : " bdw");
+ return -ENXIO;
+ }
+
+ nfit_blk = devm_kzalloc(&nd_region->dev, sizeof(*nfit_blk), GFP_KERNEL);
+ if (!nfit_blk)
+ return -ENOMEM;
+ ndbr->blk_provider_data = nfit_blk;
+
+ /* map block aperture memory */
+ nfit_blk->bdw_offset = nfit_mem->bdw->bdw_offset;
+ mmio = &nfit_blk->mmio[BDW];
+ mmio->base = nfit_spa_map(acpi_desc, nfit_mem->spa_bdw);
+ if (!mmio->base) {
+ dev_dbg(&nd_region->dev, "%s: %s failed to map bdw\n", __func__,
+ nd_dimm_name(nd_dimm));
+ return -ENOMEM;
+ }
+ mmio->size = nfit_mem->bdw->bdw_size;
+ mmio->base_offset = nfit_mem->memdev_bdw->region_spa_offset;
+ mmio->idt = nfit_mem->idt_bdw;
+ mmio->spa = nfit_mem->spa_bdw;
+ rc = nfit_blk_init_interleave(mmio, nfit_mem->idt_bdw,
+ nfit_mem->memdev_bdw->interleave_ways);
+ if (rc) {
+ dev_dbg(&nd_region->dev, "%s: %s failed to init bdw interleave\n",
+ __func__, nd_dimm_name(nd_dimm));
+ return rc;
+ }
+
+ /* map block control memory */
+ nfit_blk->cmd_offset = nfit_mem->dcr->cmd_offset;
+ nfit_blk->stat_offset = nfit_mem->dcr->status_offset;
+ mmio = &nfit_blk->mmio[DCR];
+ mmio->base = nfit_spa_map(acpi_desc, nfit_mem->spa_dcr);
+ if (!mmio->base) {
+ dev_dbg(&nd_region->dev, "%s: %s failed to map dcr\n", __func__,
+ nd_dimm_name(nd_dimm));
+ return -ENOMEM;
+ }
+ mmio->size = nfit_mem->dcr->bcw_size;
+ mmio->base_offset = nfit_mem->memdev_dcr->region_spa_offset;
+ mmio->idt = nfit_mem->idt_dcr;
+ mmio->spa = nfit_mem->spa_dcr;
+ rc = nfit_blk_init_interleave(mmio, nfit_mem->idt_dcr,
+ nfit_mem->memdev_dcr->interleave_ways);
+ if (rc) {
+ dev_dbg(&nd_region->dev, "%s: %s failed to init dcr interleave\n",
+ __func__, nd_dimm_name(nd_dimm));
+ return rc;
+ }
+
+ if (mmio->line_size == 0)
+ return 0;
+
+ if ((u32) nfit_blk->cmd_offset % mmio->line_size + 8 > mmio->line_size) {
+ dev_dbg(&nd_region->dev,
+ "cmd_offset crosses interleave boundary\n");
+ return -ENXIO;
+ } else if ((u32) nfit_blk->stat_offset % mmio->line_size + 8 > mmio->line_size) {
+ dev_dbg(&nd_region->dev,
+ "stat_offset crosses interleave boundary\n");
+ return -ENXIO;
+ }
+
+ return 0;
+}
+
+static void nd_acpi_blk_region_disable(struct nd_bus *nd_bus,
+ struct nd_blk_region *ndbr)
+{
+ struct nd_bus_descriptor *nd_desc = to_nd_desc(nd_bus);
+ struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc);
+ struct nfit_blk *nfit_blk = ndbr->blk_provider_data;
+ int i;
+
+ if (!nfit_blk)
+ return; /* never enabled */
+
+ /* auto-free BLK spa mappings */
+ for (i = 0; i < 2; i++) {
+ struct nfit_blk_mmio *mmio = &nfit_blk->mmio[i];
+
+ if (mmio->base)
+ nfit_spa_unmap(acpi_desc, mmio->spa);
+ }
+ ndbr->blk_provider_data = NULL;
+ /* devm will free nfit_blk */
+}
+
static int nd_acpi_register_region(struct acpi_nfit_desc *acpi_desc,
struct nfit_spa *nfit_spa)
{
static struct nd_mapping nd_mappings[ND_MAX_MAPPINGS];
struct acpi_nfit_spa *spa = nfit_spa->spa;
+ struct nd_blk_region_desc ndbr_desc;
+ struct nd_region_desc *ndr_desc;
struct nfit_memdev *nfit_memdev;
- struct nd_region_desc ndr_desc;
int spa_type, count = 0, rc;
struct resource res;
u16 spa_index;
@@ -885,12 +1261,13 @@ static int nd_acpi_register_region(struct acpi_nfit_desc *acpi_desc,
memset(&res, 0, sizeof(res));
memset(&nd_mappings, 0, sizeof(nd_mappings));
- memset(&ndr_desc, 0, sizeof(ndr_desc));
+ memset(&ndbr_desc, 0, sizeof(ndr_desc));
res.start = spa->spa_base;
res.end = res.start + spa->spa_length - 1;
- ndr_desc.res = &res;
- ndr_desc.provider_data = nfit_spa;
- ndr_desc.attr_groups = nd_acpi_region_attribute_groups;
+ ndr_desc = &ndbr_desc.ndr_desc;
+ ndr_desc->res = &res;
+ ndr_desc->provider_data = nfit_spa;
+ ndr_desc->attr_groups = nd_acpi_region_attribute_groups;
list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) {
struct acpi_nfit_memdev *memdev = nfit_memdev->memdev;
struct nd_mapping *nd_mapping;
@@ -926,26 +1303,29 @@ static int nd_acpi_register_region(struct acpi_nfit_desc *acpi_desc,
} else {
nd_mapping->size = nfit_mem->bdw->blk_capacity;
nd_mapping->start = nfit_mem->bdw->blk_offset;
- ndr_desc.num_lanes = nfit_mem->bdw->num_bdw;
+ ndr_desc->num_lanes = nfit_mem->bdw->num_bdw;
}
- ndr_desc.nd_mapping = nd_mapping;
- ndr_desc.num_mappings = blk_valid;
- if (!nd_blk_region_create(acpi_desc->nd_bus, &ndr_desc))
+ ndr_desc->nd_mapping = nd_mapping;
+ ndr_desc->num_mappings = blk_valid;
+ ndbr_desc.enable = nd_acpi_blk_region_enable;
+ ndbr_desc.disable = nd_acpi_blk_region_disable;
+ ndbr_desc.do_io = acpi_desc->blk_do_io;
+ if (!nd_blk_region_create(acpi_desc->nd_bus, ndr_desc))
return -ENOMEM;
}
}
- ndr_desc.nd_mapping = nd_mappings;
- ndr_desc.num_mappings = count;
- rc = nd_acpi_init_interleave_set(acpi_desc, &ndr_desc, spa);
+ ndr_desc->nd_mapping = nd_mappings;
+ ndr_desc->num_mappings = count;
+ rc = nd_acpi_init_interleave_set(acpi_desc, ndr_desc, spa);
if (rc)
return rc;
if (spa_type == NFIT_SPA_PM) {
- if (!nd_pmem_region_create(acpi_desc->nd_bus, &ndr_desc))
+ if (!nd_pmem_region_create(acpi_desc->nd_bus, ndr_desc))
return -ENOMEM;
} else if (spa_type == NFIT_SPA_VOLATILE) {
- if (!nd_volatile_region_create(acpi_desc->nd_bus, &ndr_desc))
+ if (!nd_volatile_region_create(acpi_desc->nd_bus, ndr_desc))
return -ENOMEM;
}
return 0;
@@ -972,11 +1352,14 @@ int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
acpi_size i;
int rc;
+ INIT_LIST_HEAD(&acpi_desc->spa_maps);
INIT_LIST_HEAD(&acpi_desc->spas);
INIT_LIST_HEAD(&acpi_desc->dcrs);
INIT_LIST_HEAD(&acpi_desc->bdws);
+ INIT_LIST_HEAD(&acpi_desc->idts);
INIT_LIST_HEAD(&acpi_desc->memdevs);
INIT_LIST_HEAD(&acpi_desc->dimms);
+ mutex_init(&acpi_desc->spa_map_mutex);
data = (u8 *) acpi_desc->nfit;
for (i = 0, sum = 0; i < sz; i++)
@@ -1035,6 +1418,7 @@ static int nd_acpi_add(struct acpi_device *adev)
dev_set_drvdata(dev, acpi_desc);
acpi_desc->dev = dev;
acpi_desc->nfit = (struct acpi_nfit *) tbl;
+ acpi_desc->blk_do_io = nd_acpi_blk_region_do_io;
nd_desc = &acpi_desc->nd_desc;
nd_desc->provider_name = "ACPI.NFIT";
nd_desc->ndctl = nd_acpi_ctl;
diff --git a/drivers/block/nd/acpi_nfit.h b/drivers/block/nd/acpi_nfit.h
index 2faac336c07d..a40128507551 100644
--- a/drivers/block/nd/acpi_nfit.h
+++ b/drivers/block/nd/acpi_nfit.h
@@ -226,6 +226,11 @@ struct nfit_bdw {
struct list_head list;
};
+struct nfit_idt {
+ struct acpi_nfit_idt *idt;
+ struct list_head list;
+};
+
struct nfit_memdev {
struct acpi_nfit_memdev *memdev;
struct list_head list;
@@ -236,10 +241,13 @@ struct nfit_mem {
struct nd_dimm *nd_dimm;
struct acpi_nfit_memdev *memdev_dcr;
struct acpi_nfit_memdev *memdev_pmem;
+ struct acpi_nfit_memdev *memdev_bdw;
struct acpi_nfit_dcr *dcr;
struct acpi_nfit_bdw *bdw;
struct acpi_nfit_spa *spa_dcr;
struct acpi_nfit_spa *spa_bdw;
+ struct acpi_nfit_idt *idt_dcr;
+ struct acpi_nfit_idt *idt_bdw;
struct list_head list;
struct acpi_device *adev;
unsigned long dsm_mask;
@@ -248,16 +256,55 @@ struct nfit_mem {
struct acpi_nfit_desc {
struct nd_bus_descriptor nd_desc;
struct acpi_nfit *nfit;
+ struct mutex spa_map_mutex;
+ struct list_head spa_maps;
struct list_head memdevs;
struct list_head dimms;
struct list_head spas;
struct list_head dcrs;
struct list_head bdws;
+ struct list_head idts;
struct nd_bus *nd_bus;
struct device *dev;
unsigned long dimm_dsm_force_en;
+ int (*blk_do_io)(struct nd_blk_region *ndbr, void *iobuf,
+ unsigned int len, int write, resource_size_t dpa);
+};
+
+enum nd_blk_mmio_selector {
+ BDW,
+ DCR,
+};
+
+struct nfit_blk {
+ struct nfit_blk_mmio {
+ void *base;
+ u64 size;
+ u64 base_offset;
+ u32 line_size;
+ u32 num_lines;
+ u32 table_size;
+ struct acpi_nfit_idt *idt;
+ struct acpi_nfit_spa *spa;
+ } mmio[2];
+ u64 bdw_offset; /* post interleave offset */
+ u64 stat_offset;
+ u64 cmd_offset;
};
+struct nfit_spa_mapping {
+ struct acpi_nfit_desc *acpi_desc;
+ struct acpi_nfit_spa *spa;
+ struct list_head list;
+ struct kref kref;
+ void *iomem;
+};
+
+static inline struct nfit_spa_mapping *to_spa_map(struct kref *kref)
+{
+ return container_of(kref, struct nfit_spa_mapping, kref);
+}
+
static inline struct acpi_nfit_memdev *__to_nfit_memdev(struct nfit_mem *nfit_mem)
{
if (nfit_mem->memdev_dcr)
diff --git a/drivers/block/nd/blk.c b/drivers/block/nd/blk.c
new file mode 100644
index 000000000000..8536ee8b2009
--- /dev/null
+++ b/drivers/block/nd/blk.c
@@ -0,0 +1,264 @@
+/*
+ * NVDIMM Block Window Driver
+ * Copyright (c) 2014, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+#include <linux/blkdev.h>
+#include <linux/fs.h>
+#include <linux/genhd.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/nd.h>
+#include <linux/sizes.h>
+#include "nd.h"
+
+struct nd_blk_device {
+ struct request_queue *queue;
+ struct gendisk *disk;
+ struct nd_namespace_blk *nsblk;
+ struct nd_blk_region *ndbr;
+ struct nd_io ndio;
+ size_t disk_size;
+ int id;
+};
+
+static int nd_blk_major;
+static DEFINE_IDA(nd_blk_ida);
+
+static resource_size_t to_dev_offset(struct nd_namespace_blk *nsblk,
+ resource_size_t ns_offset, unsigned int len)
+{
+ int i;
+
+ for (i = 0; i < nsblk->num_resources; i++) {
+ if (ns_offset < resource_size(nsblk->res[i])) {
+ if (ns_offset + len > resource_size(nsblk->res[i])) {
+ dev_WARN_ONCE(&nsblk->dev, 1,
+ "%s: illegal request\n", __func__);
+ return SIZE_MAX;
+ }
+ return nsblk->res[i]->start + ns_offset;
+ }
+ ns_offset -= resource_size(nsblk->res[i]);
+ }
+
+ dev_WARN_ONCE(&nsblk->dev, 1, "%s: request out of range\n", __func__);
+ return SIZE_MAX;
+}
+
+static void nd_blk_make_request(struct request_queue *q, struct bio *bio)
+{
+ struct block_device *bdev = bio->bi_bdev;
+ struct gendisk *disk = bdev->bd_disk;
+ struct nd_namespace_blk *nsblk;
+ struct nd_blk_device *blk_dev;
+ struct nd_blk_region *ndbr;
+ struct bvec_iter iter;
+ struct bio_vec bvec;
+ int err = 0, rw;
+ sector_t sector;
+
+ sector = bio->bi_iter.bi_sector;
+ if (bio_end_sector(bio) > get_capacity(disk)) {
+ err = -EIO;
+ goto out;
+ }
+
+ BUG_ON(bio->bi_rw & REQ_DISCARD);
+
+ rw = bio_data_dir(bio);
+
+ blk_dev = disk->private_data;
+ nsblk = blk_dev->nsblk;
+ ndbr = blk_dev->ndbr;
+ bio_for_each_segment(bvec, bio, iter) {
+ unsigned int len = bvec.bv_len;
+ resource_size_t dev_offset;
+ void *iobuf;
+
+ BUG_ON(len > PAGE_SIZE);
+
+ dev_offset = to_dev_offset(nsblk, sector << SECTOR_SHIFT, len);
+ if (dev_offset == SIZE_MAX) {
+ err = -EIO;
+ goto out;
+ }
+
+ iobuf = kmap_atomic(bvec.bv_page);
+ err = ndbr->do_io(ndbr, iobuf + bvec.bv_offset, len, rw, dev_offset);
+ kunmap_atomic(iobuf);
+ if (err)
+ goto out;
+
+ sector += len >> SECTOR_SHIFT;
+ }
+
+ out:
+ bio_endio(bio, err);
+}
+
+static int nd_blk_rw_bytes(struct nd_io *ndio, void *iobuf, size_t offset,
+ size_t n, unsigned long flags)
+{
+ struct nd_namespace_blk *nsblk;
+ struct nd_blk_device *blk_dev;
+ int rw = nd_data_dir(flags);
+ struct nd_blk_region *ndbr;
+ resource_size_t dev_offset;
+
+ blk_dev = container_of(ndio, typeof(*blk_dev), ndio);
+ ndbr = blk_dev->ndbr;
+ nsblk = blk_dev->nsblk;
+ dev_offset = to_dev_offset(nsblk, offset, n);
+
+ if (unlikely(offset + n > blk_dev->disk_size)) {
+ dev_WARN_ONCE(ndio->dev, 1, "%s: request out of range\n",
+ __func__);
+ return -EFAULT;
+ }
+
+ if (dev_offset == SIZE_MAX)
+ return -EIO;
+
+ return ndbr->do_io(ndbr, iobuf, n, rw, dev_offset);
+}
+
+static const struct block_device_operations nd_blk_fops = {
+ .owner = THIS_MODULE,
+};
+
+static int nd_blk_probe(struct device *dev)
+{
+ struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev);
+ struct nd_blk_device *blk_dev;
+ resource_size_t disk_size;
+ struct gendisk *disk;
+ int err;
+
+ disk_size = nd_namespace_blk_validate(nsblk);
+ if (disk_size < ND_MIN_NAMESPACE_SIZE)
+ return -ENXIO;
+
+ blk_dev = kzalloc(sizeof(*blk_dev), GFP_KERNEL);
+ if (!blk_dev)
+ return -ENOMEM;
+
+ blk_dev->id = ida_simple_get(&nd_blk_ida, 0, 0, GFP_KERNEL);
+ if (blk_dev->id < 0) {
+ err = blk_dev->id;
+ goto err_ida;
+ }
+
+ blk_dev->disk_size = disk_size;
+
+ blk_dev->queue = blk_alloc_queue(GFP_KERNEL);
+ if (!blk_dev->queue) {
+ err = -ENOMEM;
+ goto err_alloc_queue;
+ }
+
+ blk_queue_make_request(blk_dev->queue, nd_blk_make_request);
+ blk_queue_max_hw_sectors(blk_dev->queue, 1024);
+ blk_queue_bounce_limit(blk_dev->queue, BLK_BOUNCE_ANY);
+
+ disk = blk_dev->disk = alloc_disk(0);
+ if (!disk) {
+ err = -ENOMEM;
+ goto err_alloc_disk;
+ }
+
+ blk_dev->ndbr = to_blk_region(to_nd_region(nsblk->dev.parent));
+ blk_dev->nsblk = nsblk;
+
+ disk->driverfs_dev = dev;
+ disk->major = nd_blk_major;
+ disk->first_minor = 0;
+ disk->fops = &nd_blk_fops;
+ disk->private_data = blk_dev;
+ disk->queue = blk_dev->queue;
+ disk->flags = GENHD_FL_EXT_DEVT;
+ sprintf(disk->disk_name, "ndblk%d", blk_dev->id);
+ set_capacity(disk, disk_size >> SECTOR_SHIFT);
+
+ nd_bus_lock(dev);
+ dev_set_drvdata(dev, blk_dev);
+
+ add_disk(disk);
+ nd_init_ndio(&blk_dev->ndio, nd_blk_rw_bytes, dev, disk, 0);
+ nd_register_ndio(&blk_dev->ndio);
+ nd_bus_unlock(dev);
+
+ return 0;
+
+ err_alloc_disk:
+ blk_cleanup_queue(blk_dev->queue);
+ err_alloc_queue:
+ ida_simple_remove(&nd_blk_ida, blk_dev->id);
+ err_ida:
+ kfree(blk_dev);
+ return err;
+}
+
+static int nd_blk_remove(struct device *dev)
+{
+ /* FIXME: eventually need to get to nd_blk_device from struct device.
+ struct nd_namespace_io *nsio = to_nd_namespace_io(dev); */
+
+ struct nd_blk_device *blk_dev = dev_get_drvdata(dev);
+
+ nd_unregister_ndio(&blk_dev->ndio);
+ del_gendisk(blk_dev->disk);
+ put_disk(blk_dev->disk);
+ blk_cleanup_queue(blk_dev->queue);
+ ida_simple_remove(&nd_blk_ida, blk_dev->id);
+ kfree(blk_dev);
+
+ return 0;
+}
+
+static struct nd_device_driver nd_blk_driver = {
+ .probe = nd_blk_probe,
+ .remove = nd_blk_remove,
+ .drv = {
+ .name = "nd_blk",
+ },
+ .type = ND_DRIVER_NAMESPACE_BLK,
+};
+
+static int __init nd_blk_init(void)
+{
+ int rc;
+
+ rc = register_blkdev(0, "nd_blk");
+ if (rc < 0)
+ return rc;
+
+ nd_blk_major = rc;
+ rc = nd_driver_register(&nd_blk_driver);
+
+ if (rc < 0)
+ unregister_blkdev(nd_blk_major, "nd_blk");
+
+ return rc;
+}
+
+static void __exit nd_blk_exit(void)
+{
+ driver_unregister(&nd_blk_driver.drv);
+ unregister_blkdev(nd_blk_major, "nd_blk");
+}
+
+MODULE_AUTHOR("Ross Zwisler <[email protected]>");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS_ND_DEVICE(ND_DEVICE_NAMESPACE_BLK);
+module_init(nd_blk_init);
+module_exit(nd_blk_exit);
diff --git a/drivers/block/nd/libnd.h b/drivers/block/nd/libnd.h
index e188840ed2b9..40373caf6d04 100644
--- a/drivers/block/nd/libnd.h
+++ b/drivers/block/nd/libnd.h
@@ -80,6 +80,16 @@ struct nd_region_desc {
};
struct nd_bus;
+struct nd_blk_region;
+struct nd_blk_region_desc {
+ int (*enable)(struct nd_bus *nd_bus, struct nd_blk_region *ndbr);
+ void (*disable)(struct nd_bus *nd_bus, struct nd_blk_region *ndbr);
+ int (*do_io)(struct nd_blk_region *ndbr, void *iobuf, unsigned int len,
+ int write, resource_size_t dpa);
+ struct nd_region_desc ndr_desc;
+};
+
+struct nd_bus;
struct nd_bus *__nd_bus_register(struct device *parent,
struct nd_bus_descriptor *nfit_desc, struct module *module);
#define nd_bus_register(parent, desc) \
@@ -91,7 +101,6 @@ struct nd_region *to_nd_region(struct device *dev);
struct nd_bus_descriptor *to_nd_desc(struct nd_bus *nd_bus);
const char *nd_dimm_name(struct nd_dimm *nd_dimm);
void *nd_dimm_provider_data(struct nd_dimm *nd_dimm);
-void *nd_region_provider_data(struct nd_region *nd_region);
struct nd_dimm *nd_dimm_create(struct nd_bus *nd_bus, void *provider_data,
const struct attribute_group **groups, unsigned long flags,
unsigned long *dsm_mask);
diff --git a/drivers/block/nd/namespace_devs.c b/drivers/block/nd/namespace_devs.c
index 4dc499758636..2a4d28867c58 100644
--- a/drivers/block/nd/namespace_devs.c
+++ b/drivers/block/nd/namespace_devs.c
@@ -151,6 +151,53 @@ static resource_size_t nd_namespace_blk_size(struct nd_namespace_blk *nsblk)
return size;
}
+resource_size_t nd_namespace_blk_validate(struct nd_namespace_blk *nsblk)
+{
+ struct nd_region *nd_region = to_nd_region(nsblk->dev.parent);
+ struct nd_mapping *nd_mapping = &nd_region->mapping[0];
+ struct nd_dimm_drvdata *ndd = to_ndd(nd_mapping);
+ struct nd_label_id label_id;
+ struct resource *res;
+ int count, i;
+
+ if (!nsblk->uuid || !nsblk->lbasize)
+ return 0;
+
+ count = 0;
+ nd_label_gen_id(&label_id, nsblk->uuid, NSLABEL_FLAG_LOCAL);
+ for_each_dpa_resource(ndd, res) {
+ if (strcmp(res->name, label_id.id) != 0)
+ continue;
+ /*
+ * Resources with unacknoweldged adjustments indicate a
+ * failure to update labels
+ */
+ if (res->flags & DPA_RESOURCE_ADJUSTED)
+ return 0;
+ count++;
+ }
+
+ /* These values match after a successful label update */
+ if (count != nsblk->num_resources)
+ return 0;
+
+ for (i = 0; i < nsblk->num_resources; i++) {
+ struct resource *found = NULL;
+
+ for_each_dpa_resource(ndd, res)
+ if (res == nsblk->res[i]) {
+ found = res;
+ break;
+ }
+ /* stale resource */
+ if (!found)
+ return 0;
+ }
+
+ return nd_namespace_blk_size(nsblk);
+}
+EXPORT_SYMBOL(nd_namespace_blk_validate);
+
static int nd_namespace_label_update(struct nd_region *nd_region, struct device *dev)
{
dev_WARN_ONCE(dev, dev->driver,
diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
index 68e9ec824dc8..a9eb4da240a6 100644
--- a/drivers/block/nd/nd-private.h
+++ b/drivers/block/nd/nd-private.h
@@ -22,7 +22,6 @@ extern struct list_head nd_bus_list;
extern struct mutex nd_bus_list_mutex;
extern int nd_dimm_major;
-struct block_device;
struct nd_io_claim;
struct nd_btt;
struct nd_io;
@@ -50,8 +49,8 @@ struct nd_dimm {
struct nd_io *ndio_lookup(struct nd_bus *nd_bus, const char *diskname);
bool is_nd_dimm(struct device *dev);
-bool is_nd_blk(struct device *dev);
bool is_nd_pmem(struct device *dev);
+bool is_nd_blk(struct device *dev);
#if IS_ENABLED(CONFIG_ND_BTT_DEVS)
bool is_nd_btt(struct device *dev);
struct nd_btt *nd_btt_create(struct nd_bus *nd_bus);
diff --git a/drivers/block/nd/nd.h b/drivers/block/nd/nd.h
index a29fb7409925..9c5a48fce0f2 100644
--- a/drivers/block/nd/nd.h
+++ b/drivers/block/nd/nd.h
@@ -113,6 +113,20 @@ struct nd_region {
struct nd_mapping mapping[0];
};
+struct nd_blk_region {
+ int (*enable)(struct nd_bus *nd_bus, struct nd_blk_region *ndbr);
+ void (*disable)(struct nd_bus *nd_bus, struct nd_blk_region *ndbr);
+ int (*do_io)(struct nd_blk_region *ndbr, void *iobuf, unsigned int len,
+ int write, resource_size_t dpa);
+ void *blk_provider_data;
+ struct nd_region nd_region;
+};
+
+static inline struct nd_blk_region *to_blk_region(struct nd_region *nd_region)
+{
+ return container_of(nd_region, struct nd_blk_region, nd_region);
+}
+
/*
* Lookup next in the repeating sequence of 01, 10, and 11.
*/
@@ -242,4 +256,6 @@ void nd_bus_unlock(struct device *dev);
bool is_nd_bus_locked(struct device *dev);
int nd_label_reserve_dpa(struct nd_dimm_drvdata *ndd);
void nd_dimm_free_dpa(struct nd_dimm_drvdata *ndd, struct resource *res);
+int nd_blk_region_init(struct nd_region *nd_region);
+resource_size_t nd_namespace_blk_validate(struct nd_namespace_blk *nsblk);
#endif /* __ND_H__ */
diff --git a/drivers/block/nd/region.c b/drivers/block/nd/region.c
index dd5a885cea11..442fbd25631c 100644
--- a/drivers/block/nd/region.c
+++ b/drivers/block/nd/region.c
@@ -83,11 +83,15 @@ EXPORT_SYMBOL(nd_region_release_lane);
static int nd_region_probe(struct device *dev)
{
- int err;
+ int err, rc;
struct nd_region_namespaces *num_ns;
struct nd_region *nd_region = to_nd_region(dev);
- int rc = nd_region_register_namespaces(nd_region, &err);
+ rc = nd_blk_region_init(nd_region);
+ if (rc)
+ return rc;
+
+ rc = nd_region_register_namespaces(nd_region, &err);
num_ns = devm_kzalloc(dev, sizeof(*num_ns), GFP_KERNEL);
if (!num_ns)
return -ENOMEM;
diff --git a/drivers/block/nd/region_devs.c b/drivers/block/nd/region_devs.c
index 268d9ef67f9c..b1fb63d0deb9 100644
--- a/drivers/block/nd/region_devs.c
+++ b/drivers/block/nd/region_devs.c
@@ -11,6 +11,7 @@
* General Public License for more details.
*/
#include <linux/scatterlist.h>
+#include <linux/highmem.h>
#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/sort.h>
@@ -33,7 +34,10 @@ static void nd_region_release(struct device *dev)
put_device(&nd_dimm->dev);
}
ida_simple_remove(®ion_ida, nd_region->id);
- kfree(nd_region);
+ if (is_nd_blk(dev))
+ kfree(to_blk_region(nd_region));
+ else
+ kfree(nd_region);
}
static struct device_type nd_blk_device_type = {
@@ -339,27 +343,33 @@ u64 nd_region_interleave_set_cookie(struct nd_region *nd_region)
/*
* Upon successful probe/remove, take/release a reference on the
- * associated interleave set (if present)
+ * associated dimms in the interleave set, on successful probe of a BLK
+ * namespace check if we need a new seed, and on remove or failed probe
+ * of a BLK region notify the provider to disable the region.
*/
static void nd_region_notify_driver_action(struct nd_bus *nd_bus,
struct device *dev, int rc, bool probe)
{
- if (rc)
- return;
-
if (is_nd_pmem(dev) || is_nd_blk(dev)) {
struct nd_region *nd_region = to_nd_region(dev);
+ struct nd_blk_region *nd_blk_region;
int i;
for (i = 0; i < nd_region->ndr_mappings; i++) {
struct nd_mapping *nd_mapping = &nd_region->mapping[i];
struct nd_dimm *nd_dimm = nd_mapping->nd_dimm;
- if (probe)
+ if (probe && rc == 0)
atomic_inc(&nd_dimm->busy);
- else
+ else if (!probe)
atomic_dec(&nd_dimm->busy);
}
+
+ if (is_nd_pmem(dev) || (probe && rc == 0))
+ return;
+
+ nd_blk_region = to_blk_region(nd_region);
+ nd_blk_region->disable(nd_bus, nd_blk_region);
} else if (dev->parent && is_nd_blk(dev->parent) && probe && rc == 0) {
struct nd_region *nd_region = to_nd_region(dev->parent);
@@ -503,11 +513,21 @@ struct attribute_group nd_mapping_attribute_group = {
};
EXPORT_SYMBOL_GPL(nd_mapping_attribute_group);
-void *nd_region_provider_data(struct nd_region *nd_region)
+int nd_blk_region_init(struct nd_region *nd_region)
{
- return nd_region->provider_data;
+ struct nd_blk_region *ndbr = to_blk_region(nd_region);
+ struct nd_bus *nd_bus = walk_to_nd_bus(&nd_region->dev);
+
+ if (!is_nd_blk(&nd_region->dev))
+ return 0;
+
+ if (nd_region->ndr_mappings < 1) {
+ dev_err(&nd_region->dev, "invalid BLK region\n");
+ return -ENXIO;
+ }
+
+ return ndbr->enable(nd_bus, ndbr);
}
-EXPORT_SYMBOL_GPL(nd_region_provider_data);
static noinline struct nd_region *nd_region_create(struct nd_bus *nd_bus,
struct nd_region_desc *ndr_desc, struct device_type *dev_type)
@@ -529,9 +549,28 @@ static noinline struct nd_region *nd_region_create(struct nd_bus *nd_bus,
}
}
- nd_region = kzalloc(sizeof(struct nd_region)
- + sizeof(struct nd_mapping) * ndr_desc->num_mappings,
- GFP_KERNEL);
+ if (dev_type == &nd_blk_device_type) {
+ struct nd_blk_region_desc *ndbr_desc;
+ struct nd_blk_region *ndbr;
+
+ ndbr_desc = container_of(ndr_desc, typeof(*ndbr_desc), ndr_desc);
+ ndbr = kzalloc(sizeof(*ndbr) + sizeof(struct nd_mapping)
+ * ndr_desc->num_mappings,
+ GFP_KERNEL);
+ if (ndbr) {
+ nd_region = &ndbr->nd_region;
+ ndbr->enable = ndbr_desc->enable;
+ ndbr->disable = ndbr_desc->disable;
+ ndbr->do_io = ndbr_desc->do_io;
+ } else
+ nd_region = NULL;
+ } else {
+ nd_region = kzalloc(sizeof(struct nd_region)
+ + sizeof(struct nd_mapping)
+ * ndr_desc->num_mappings,
+ GFP_KERNEL);
+ }
+
if (!nd_region)
return NULL;
nd_region->id = ida_simple_get(®ion_ida, 0, 0, GFP_KERNEL);
diff --git a/drivers/block/nd/test/nfit.c b/drivers/block/nd/test/nfit.c
index 50916f0ca901..ea4b2063f2c3 100644
--- a/drivers/block/nd/test/nfit.c
+++ b/drivers/block/nd/test/nfit.c
@@ -13,6 +13,7 @@
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/platform_device.h>
#include <linux/dma-mapping.h>
+#include <linux/device.h>
#include <linux/module.h>
#include <linux/ndctl.h>
#include <linux/sizes.h>
@@ -21,6 +22,7 @@
#include "../acpi_nfit.h"
#include "../libnd.h"
+#include "../nd.h"
/*
* Generate an NFIT table to describe the following topology:
@@ -907,6 +909,32 @@ static void nfit_test1_setup(struct nfit_test *t)
nfit->checksum = nfit_checksum(nfit_buf, size);
}
+static int nfit_test_blk_do_io(struct nd_blk_region *ndbr, void *iobuf,
+ unsigned int len, int rw, resource_size_t dpa)
+{
+ struct nfit_blk *nfit_blk = ndbr->blk_provider_data;
+ struct nfit_blk_mmio *mmio = &nfit_blk->mmio[BDW];
+ struct nd_region *nd_region = &ndbr->nd_region;
+ struct nfit_test_resource *nfit_res;
+ unsigned int bw;
+
+ nfit_res = nfit_test_lookup((unsigned long) mmio->base);
+ if (!nfit_res) {
+ dev_WARN_ONCE(&nd_region->dev, 1, "no test resource\n");
+ return -EIO;
+ }
+ dev_vdbg(&nd_region->dev, "%s: base: %p offset: %pa\n",
+ __func__, mmio->base, &dpa);
+ bw = nd_region_acquire_lane(nd_region);
+ if (rw)
+ memcpy(nfit_res->buf + dpa, iobuf, len);
+ else
+ memcpy(iobuf, nfit_res->buf + dpa, len);
+ nd_region_release_lane(nd_region, bw);
+
+ return 0;
+}
+
extern const struct attribute_group *nd_acpi_attribute_groups[];
static int nfit_test_probe(struct platform_device *pdev)
@@ -957,6 +985,7 @@ static int nfit_test_probe(struct platform_device *pdev)
acpi_desc = &nfit_test->acpi_desc;
acpi_desc->dev = &pdev->dev;
acpi_desc->nfit = nfit_test->nfit_buf;
+ acpi_desc->blk_do_io = nfit_test_blk_do_io;
nd_desc = &acpi_desc->nd_desc;
nd_desc->attr_groups = nd_acpi_attribute_groups;
acpi_desc->nd_bus = nd_bus_register(&pdev->dev, nd_desc);
diff --git a/drivers/block/nd/test/nfit_test.h b/drivers/block/nd/test/nfit_test.h
index 7b071478eb94..30423b5b4b6f 100644
--- a/drivers/block/nd/test/nfit_test.h
+++ b/drivers/block/nd/test/nfit_test.h
@@ -21,6 +21,6 @@ struct nfit_test_resource {
};
typedef struct nfit_test_resource *(*nfit_test_lookup_fn)(resource_size_t);
-void nfit_test_setup(nfit_test_lookup_fn fn);
+void nfit_test_setup(nfit_test_lookup_fn lookup);
void nfit_test_teardown(void);
#endif
On Tue, Apr 28, 2015 at 11:24 AM, Dan Williams <[email protected]> wrote:
> ACPI 6.0 formalizes e820-type-7 and efi-type-14 as persistent memory.
> Mark it "reserved" and allow it to be claimed by a persistent memory
> device driver.
>
> This definition is in addition to the Linux kernel's existing type-12
> definition that was recently added in support of shipping platforms with
> NVDIMM support that predate ACPI 6.0 (which now classifies type-12 as
> OEM reserved). We may choose to exploit this wealth of definitions for
> NVDIMMs to differentiate E820_PRAM (type-12) from E820_PMEM (type-7).
> One potential differentiation is that PMEM is not backed by struct page
> by default in contrast to PRAM. For now, they are effectively treated
> as aliases by the mm.
>
> Note, /proc/iomem can be consulted for differentiating legacy
> "Persistent RAM" E820_PRAM vs standard "Persistent I/O Memory"
> E820_PMEM.
>
> Cc: Boaz Harrosh <[email protected]>
> Cc: Ingo Molnar <[email protected]>
> Cc: Christoph Hellwig <[email protected]>
> Cc: Andrew Morton <[email protected]>
> Cc: Borislav Petkov <[email protected]>
> Cc: H. Peter Anvin <[email protected]>
> Cc: Jens Axboe <[email protected]>
> Cc: Linus Torvalds <[email protected]>
> Cc: Matthew Wilcox <[email protected]>
> Cc: Thomas Gleixner <[email protected]>
> Acked-by: Andy Lutomirski <[email protected]>
> Reviewed-by: Ross Zwisler <[email protected]>
> Signed-off-by: Dan Williams <[email protected]>
> ---
> arch/arm64/kernel/efi.c | 1 +
> arch/ia64/kernel/efi.c | 4 ++++
> arch/x86/boot/compressed/eboot.c | 4 ++++
> arch/x86/include/uapi/asm/e820.h | 1 +
> arch/x86/kernel/e820.c | 26 +++++++++++++++++++++++---
> arch/x86/platform/efi/efi.c | 3 +++
> include/linux/efi.h | 3 ++-
> 7 files changed, 38 insertions(+), 4 deletions(-)
>
> diff --git a/arch/arm64/kernel/efi.c b/arch/arm64/kernel/efi.c
> index ab21e0d58278..9d4aa18f2a82 100644
> --- a/arch/arm64/kernel/efi.c
> +++ b/arch/arm64/kernel/efi.c
> @@ -158,6 +158,7 @@ static __init int is_reserve_region(efi_memory_desc_t *md)
> case EFI_BOOT_SERVICES_CODE:
> case EFI_BOOT_SERVICES_DATA:
> case EFI_CONVENTIONAL_MEMORY:
> + case EFI_PERSISTENT_MEMORY:
> return 0;
> default:
> break;
> diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c
> index c52d7540dc05..9028bc268cd7 100644
> --- a/arch/ia64/kernel/efi.c
> +++ b/arch/ia64/kernel/efi.c
> @@ -1223,6 +1223,10 @@ efi_initialize_iomem_resources(struct resource *code_resource,
> flags |= IORESOURCE_DISABLED;
> break;
>
> + case EFI_PERSISTENT_MEMORY:
> + name = "persistent";
> + break;
> +
> case EFI_RESERVED_TYPE:
> case EFI_RUNTIME_SERVICES_CODE:
> case EFI_RUNTIME_SERVICES_DATA:
> diff --git a/arch/x86/boot/compressed/eboot.c b/arch/x86/boot/compressed/eboot.c
> index ef17683484e9..dde5bf7726f4 100644
> --- a/arch/x86/boot/compressed/eboot.c
> +++ b/arch/x86/boot/compressed/eboot.c
> @@ -1222,6 +1222,10 @@ static efi_status_t setup_e820(struct boot_params *params,
> e820_type = E820_NVS;
> break;
>
> + case EFI_PERSISTENT_MEMORY:
> + e820_type = E820_PMEM;
> + break;
> +
> default:
> continue;
> }
> diff --git a/arch/x86/include/uapi/asm/e820.h b/arch/x86/include/uapi/asm/e820.h
> index 960a8a9dc4ab..0f457e6eab18 100644
> --- a/arch/x86/include/uapi/asm/e820.h
> +++ b/arch/x86/include/uapi/asm/e820.h
> @@ -32,6 +32,7 @@
> #define E820_ACPI 3
> #define E820_NVS 4
> #define E820_UNUSABLE 5
> +#define E820_PMEM 7
>
> /*
> * This is a non-standardized way to represent ADR or NVDIMM regions that
> diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
> index 11cc7d54ec3f..d38b53a7e9b2 100644
> --- a/arch/x86/kernel/e820.c
> +++ b/arch/x86/kernel/e820.c
> @@ -149,6 +149,7 @@ static void __init e820_print_type(u32 type)
> case E820_UNUSABLE:
> printk(KERN_CONT "unusable");
> break;
> + case E820_PMEM:
> case E820_PRAM:
> printk(KERN_CONT "persistent (type %u)", type);
> break;
I'd kind of like to make it more clear what's going on here. It
doesn't help that the spec chose poor names.
How about "NVDIMM physical aperture" for E820_PMEM and "legacy
persistent RAM" for E820_PRAM?
Otherwise this looks generaly sensible, although I don't really
understand why e820_type_to_string and e820_print_type are different.
--Andy
On Tue, Apr 28, 2015 at 11:24 AM, Dan Williams <[email protected]> wrote:
> Changes since v1 [1]: Incorporates feedback received prior to April 24.
>
> 1/ Ingo said [2]:
>
> "So why on earth is this whole concept and the naming itself
> ('drivers/block/nd/' stands for 'NFIT Defined', apparently)
> revolving around a specific 'firmware' mindset and revolving
> around specific, weirdly named, overly complicated looking
> firmware interfaces that come with their own new weird
> glossary??"
>
> Indeed, we of course consulted the NFIT specification to determine
> the shape of the sub-system, but then let its terms and data
> structures permeate too deep into the implementation. That is fixed
> now with all NFIT specifics factored out into acpi.c. The NFIT is no
> longer required reading to review libnd. Only three concepts are
> needed:
>
> i/ PMEM - contiguous memory range where cpu stores are
> persistent once they are flushed through the memory
> controller.
>
> ii/ BLK - mmio apertures (sliding windows) that can be
> programmed to access an aperture's-worth of persistent
> media at a time.
>
> iii/ DPA - "dimm-physical-address", address space local to a
> dimm. A dimm may provide both PMEM-mode and BLK-mode
> access to a range of DPA. libnd manages allocation of DPA
> to either PMEM or BLK-namespaces to resolve this aliasing.
Mostly for my understanding: is there a name for "address relative to
the address lines on the DIMM"? That is, a DIMM that exposes 8 GB of
apparent physical memory, possibly interleaved, broken up, or weirdly
remapped by the memory controller, would still have addresses between
0 and 8 GB. Some of those might be PMEM windows, some might be MMIO,
some might be BLK apertures, etc.
IIUC "DPA" refers to actual addressable storage, not this type of address?
--Andy
On Tue, Apr 28, 2015 at 1:49 PM, Andy Lutomirski <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 11:24 AM, Dan Williams <[email protected]> wrote:
>> diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
>> index 11cc7d54ec3f..d38b53a7e9b2 100644
>> --- a/arch/x86/kernel/e820.c
>> +++ b/arch/x86/kernel/e820.c
>> @@ -149,6 +149,7 @@ static void __init e820_print_type(u32 type)
>> case E820_UNUSABLE:
>> printk(KERN_CONT "unusable");
>> break;
>> + case E820_PMEM:
>> case E820_PRAM:
>> printk(KERN_CONT "persistent (type %u)", type);
>> break;
>
> I'd kind of like to make it more clear what's going on here. It
> doesn't help that the spec chose poor names.
>
> How about "NVDIMM physical aperture" for E820_PMEM and "legacy
> persistent RAM" for E820_PRAM?
The term "aperture" to me implies this BLK (mmio-windowed) mode of
accessing persistent media that the NFIT specification introduces. In
fact, those ranges are mapped E820_RESERVED. E820_PMEM really is a
memory range that happens to be persistent.
> Otherwise this looks generaly sensible, although I don't really
> understand why e820_type_to_string and e820_print_type are different.
e820_type_to_string() appears in /proc/iomem and seems to afford
being more descriptive than e820_print_type() that just scrolls by in
dmesg, but I'm just guessing.
On Tue, Apr 28, 2015 at 1:52 PM, Andy Lutomirski <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 11:24 AM, Dan Williams <[email protected]> wrote:
>> Changes since v1 [1]: Incorporates feedback received prior to April 24.
>>
>> 1/ Ingo said [2]:
>>
>> "So why on earth is this whole concept and the naming itself
>> ('drivers/block/nd/' stands for 'NFIT Defined', apparently)
>> revolving around a specific 'firmware' mindset and revolving
>> around specific, weirdly named, overly complicated looking
>> firmware interfaces that come with their own new weird
>> glossary??"
>>
>> Indeed, we of course consulted the NFIT specification to determine
>> the shape of the sub-system, but then let its terms and data
>> structures permeate too deep into the implementation. That is fixed
>> now with all NFIT specifics factored out into acpi.c. The NFIT is no
>> longer required reading to review libnd. Only three concepts are
>> needed:
>>
>> i/ PMEM - contiguous memory range where cpu stores are
>> persistent once they are flushed through the memory
>> controller.
>>
>> ii/ BLK - mmio apertures (sliding windows) that can be
>> programmed to access an aperture's-worth of persistent
>> media at a time.
>>
>> iii/ DPA - "dimm-physical-address", address space local to a
>> dimm. A dimm may provide both PMEM-mode and BLK-mode
>> access to a range of DPA. libnd manages allocation of DPA
>> to either PMEM or BLK-namespaces to resolve this aliasing.
>
> Mostly for my understanding: is there a name for "address relative to
> the address lines on the DIMM"? That is, a DIMM that exposes 8 GB of
> apparent physical memory, possibly interleaved, broken up, or weirdly
> remapped by the memory controller, would still have addresses between
> 0 and 8 GB. Some of those might be PMEM windows, some might be MMIO,
> some might be BLK apertures, etc.
>
> IIUC "DPA" refers to actual addressable storage, not this type of address?
No, DPA is exactly as you describe above. You can't directly access
it except through a PMEM mapping (possibly interleaved with DPA from
other DIMMs) or a BLK aperture (mmio window into DPA).
On Tue, Apr 28, 2015 at 11:25 AM, Dan Williams <[email protected]> wrote:
> nd_pmem attaches to persistent memory regions and namespaces emitted by
> the nd subsystem, and, same as the original pmem driver, presents the
> system-physical-address range as a block device.
>
> The existing e820-type-12 to pmem setup is converted to a full libnd bus
> that emits an nd_namespace_io device.
Now that I think I understand what's going on, can I propose alternate
help-text wording?
[reordered the quotes]
> +config BLK_DEV_PMEM
> + tristate "PMEM: Persistent memory block device support"
> + depends on LIBND
> + default LIBND
> + help
> + Memory ranges for PMEM are described by either an NFIT
> + (NVDIMM Firmware Interface Table, see CONFIG_NFIT_ACPI), a
> + non-standard OEM-specific E820 memory type (type-12, see
> + CONFIG_X86_PMEM_LEGACY), or it is manually specified by the
> + 'memmap=nn[KMG]!ss[KMG]' kernel command line (see
> + Documentation/kernel-parameters.txt). This driver converts
> + these persistent memory ranges into block devices that are
> + capable of DAX (direct-access) file system mappings. See
> + Documentation/blockdev/nd.txt for more details.
How about something like:
"This driver exposes memory devices that expose byte-addressable
nonvolatile storage as block devices. This type of block device
supports direct access (i.e. DAX) if an appropriate filesystem is
used.
This driver requires a libnd driver that supports the memory device being used."
Then there would be a sub-menu with the legacy and NFIT drivers. (Or,
if this isn't actually configurable that way, then there could be a
reference to what needs to be configured to make this work.)
> +config ND_E820
> + tristate "E820: Support the E820-type-12 PMEM convention"
> + depends on X86_PMEM_LEGACY
> + default m if X86_PMEM_LEGACY
> + select LIBND
> + help
> + Prior to ACPI 6 some platforms advertised peristent memory
> + via type-12 e820 memory ranges. Create a libnd bus and
> + attach an instance of the pmem driver to these ranges.
> +
How about something like:
"This driver allows libnd to work with legacy, pre-ACPI 6 NVDIMMs.
This enables such devices to be exposed as block devices using PMEM.
The legacy NVDIMM interface is problematic. This driver will not work
if you boot using UEFI, and some NVDIMMs and motherboards that work
with this driver may require proprietary code in order to work
reliably."
--Andy
On Tue, Apr 28, 2015 at 1:57 PM, Dan Williams <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 1:49 PM, Andy Lutomirski <[email protected]> wrote:
>> On Tue, Apr 28, 2015 at 11:24 AM, Dan Williams <[email protected]> wrote:
>>> diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
>>> index 11cc7d54ec3f..d38b53a7e9b2 100644
>>> --- a/arch/x86/kernel/e820.c
>>> +++ b/arch/x86/kernel/e820.c
>>> @@ -149,6 +149,7 @@ static void __init e820_print_type(u32 type)
>>> case E820_UNUSABLE:
>>> printk(KERN_CONT "unusable");
>>> break;
>>> + case E820_PMEM:
>>> case E820_PRAM:
>>> printk(KERN_CONT "persistent (type %u)", type);
>>> break;
>>
>> I'd kind of like to make it more clear what's going on here. It
>> doesn't help that the spec chose poor names.
>>
>> How about "NVDIMM physical aperture" for E820_PMEM and "legacy
>> persistent RAM" for E820_PRAM?
>
> The term "aperture" to me implies this BLK (mmio-windowed) mode of
> accessing persistent media that the NFIT specification introduces. In
> fact, those ranges are mapped E820_RESERVED. E820_PMEM really is a
> memory range that happens to be persistent.
Oh, I missed that. Yuck. What happens when you repartition one of
these things? (Can you even do that?)
>
>> Otherwise this looks generaly sensible, although I don't really
>> understand why e820_type_to_string and e820_print_type are different.
>
> e820_type_to_string() appears in /proc/iomem and seems to afford
> being more descriptive than e820_print_type() that just scrolls by in
> dmesg, but I'm just guessing.
Can we change that?
--
Andy Lutomirski
AMA Capital Management, LLC
On Tue, Apr 28, 2015 at 1:59 PM, Dan Williams <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 1:52 PM, Andy Lutomirski <[email protected]> wrote:
>> On Tue, Apr 28, 2015 at 11:24 AM, Dan Williams <[email protected]> wrote:
>>> Changes since v1 [1]: Incorporates feedback received prior to April 24.
>>>
>>> 1/ Ingo said [2]:
>>>
>>> "So why on earth is this whole concept and the naming itself
>>> ('drivers/block/nd/' stands for 'NFIT Defined', apparently)
>>> revolving around a specific 'firmware' mindset and revolving
>>> around specific, weirdly named, overly complicated looking
>>> firmware interfaces that come with their own new weird
>>> glossary??"
>>>
>>> Indeed, we of course consulted the NFIT specification to determine
>>> the shape of the sub-system, but then let its terms and data
>>> structures permeate too deep into the implementation. That is fixed
>>> now with all NFIT specifics factored out into acpi.c. The NFIT is no
>>> longer required reading to review libnd. Only three concepts are
>>> needed:
>>>
>>> i/ PMEM - contiguous memory range where cpu stores are
>>> persistent once they are flushed through the memory
>>> controller.
>>>
>>> ii/ BLK - mmio apertures (sliding windows) that can be
>>> programmed to access an aperture's-worth of persistent
>>> media at a time.
>>>
>>> iii/ DPA - "dimm-physical-address", address space local to a
>>> dimm. A dimm may provide both PMEM-mode and BLK-mode
>>> access to a range of DPA. libnd manages allocation of DPA
>>> to either PMEM or BLK-namespaces to resolve this aliasing.
>>
>> Mostly for my understanding: is there a name for "address relative to
>> the address lines on the DIMM"? That is, a DIMM that exposes 8 GB of
>> apparent physical memory, possibly interleaved, broken up, or weirdly
>> remapped by the memory controller, would still have addresses between
>> 0 and 8 GB. Some of those might be PMEM windows, some might be MMIO,
>> some might be BLK apertures, etc.
>>
>> IIUC "DPA" refers to actual addressable storage, not this type of address?
>
> No, DPA is exactly as you describe above. You can't directly access
> it except through a PMEM mapping (possibly interleaved with DPA from
> other DIMMs) or a BLK aperture (mmio window into DPA).
So the thing I'm describing has no name, then? Oh, well.
--Andy
On Tue, Apr 28, 2015 at 11:26 AM, Dan Williams <[email protected]> wrote:
> From: Ross Zwisler <[email protected]>
>
> The libnd implementation handles allocating dimm address space (DPA)
> between PMEM and BLK mode interfaces. After DPA has been allocated from
> a BLK-region to a BLK-namespace the nd_blk driver attaches to handle I/O
> as a struct bio based block device. Unlike PMEM, BLK is required to
> handle platform specific details like mmio register formats and memory
> controller interleave. For this reason the libnd generic nd_blk driver
> calls back into the bus provider to carry out the I/O.
>
> This initial implementation handles the BLK interface defined by the
> ACPI 6 NFIT [1] and the NVDIMM DSM Interface Example [2] composed from
> DCR (dimm control region), BDW (block data window), IDT (interleave
> descriptor) NFIT structures and the hardware register format.
> [1]: http://www.uefi.org/sites/default/files/resources/ACPI_6.0.pdf
> [2]: http://pmem.io/documents/NVDIMM_DSM_Interface_Example.pdf
>
> Cc: Andy Lutomirski <[email protected]>
> Cc: Boaz Harrosh <[email protected]>
> Cc: H. Peter Anvin <[email protected]>
> Cc: Jens Axboe <[email protected]>
> Cc: Ingo Molnar <[email protected]>
> Cc: Christoph Hellwig <[email protected]>
> Signed-off-by: Ross Zwisler <[email protected]>
> Signed-off-by: Dan Williams <[email protected]>
> ---
> drivers/block/nd/Kconfig | 12 +
> drivers/block/nd/Makefile | 3
> drivers/block/nd/acpi.c | 422 +++++++++++++++++++++++++++++++++++--
> drivers/block/nd/acpi_nfit.h | 47 ++++
> drivers/block/nd/blk.c | 264 +++++++++++++++++++++++
> drivers/block/nd/libnd.h | 11 +
> drivers/block/nd/namespace_devs.c | 47 ++++
> drivers/block/nd/nd-private.h | 3
> drivers/block/nd/nd.h | 16 +
> drivers/block/nd/region.c | 8 +
> drivers/block/nd/region_devs.c | 65 +++++-
> drivers/block/nd/test/nfit.c | 29 +++
> drivers/block/nd/test/nfit_test.h | 2
> 13 files changed, 891 insertions(+), 38 deletions(-)
> create mode 100644 drivers/block/nd/blk.c
>
> diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
> index 612bf2b14283..bac4290129fc 100644
> --- a/drivers/block/nd/Kconfig
> +++ b/drivers/block/nd/Kconfig
> @@ -95,6 +95,18 @@ config BLK_DEV_PMEM
>
> Say Y if you want to use a NVDIMM described by ACPI, E820, etc...
>
> +config ND_BLK
> + tristate "BLK: Block data window (aperture) device support"
> + depends on LIBND
> + default ND_ACPI
> + help
> + This driver performs I/O using a set of mmio windows on a
> + dimm. The set of apertures will all access the one DIMM.
> + Multiple windows allow multiple threads to have a different
> + portions of the dimm open at one time.
> +
> + Say Y if you want to use a NVDIMM with BLK-mode capability
> +
This describes how it works, not what it is. How about:
This driver exposes NVDIMM BLK regions as block devices. BLK regions
are regions of NVDIMM storage that are sector-addressable, not
byte-addressible, and do not support DAX.
Say Y if you want to...
--Andy
> -----Original Message-----
> From: Linux-nvdimm [mailto:[email protected]] On Behalf Of
> Dan Williams
> Sent: Tuesday, April 28, 2015 1:24 PM
> To: [email protected]
> Cc: Neil Brown; Dave Chinner; H. Peter Anvin; Christoph Hellwig; Rafael J.
> Wysocki; Robert Moore; Ingo Molnar; [email protected]; Jens Axboe;
> Borislav Petkov; Thomas Gleixner; Greg KH; [email protected];
> Andy Lutomirski; Andrew Morton; Linus Torvalds
> Subject: [Linux-nvdimm] [PATCH v2 00/20] libnd: non-volatile memory device
> support
>
> Changes since v1 [1]: Incorporates feedback received prior to April 24.
Here are some comments on the sysfs properties reported for a pmem device.
They are based on v1, but I don't think v2 changes anything.
1. This confuses lsblk (part of util-linux):
/sys/block/pmem0/device/type:4
lsblk shows:
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
pmem0 251:0 0 8G 0 worm
pmem1 251:16 0 8G 0 worm
pmem2 251:32 0 8G 0 worm
pmem3 251:48 0 8G 0 worm
pmem4 251:64 0 8G 0 worm
pmem5 251:80 0 8G 0 worm
pmem6 251:96 0 8G 0 worm
pmem7 251:112 0 8G 0 worm
lsblk's blkdev_scsi_type_to_name() considers 4 to mean
SCSI_TYPE_WORM (write once read many ... used for certain optical
and tape drives).
I'm not sure what nd and pmem are doing to result in that value.
2. To avoid confusing software trying to detect fast storage vs.
slow storage devices via sysfs, this value should be 0:
/sys/block/pmem0/queue/rotational:1
That can be done by adding this shortly after the blk_alloc_queue call:
queue_flag_set_unlocked(QUEUE_FLAG_NONROT, pmem->pmem_queue);
3. Is there any reason to have a 512 KiB limit on the transfer
length?
/sys/block/pmem0/queue/max_hw_sectors_kb:512
That is from:
blk_queue_max_hw_sectors(pmem->pmem_queue, 1024);
4. These are read-writeable, but IOs never reach a queue, so
the queue size is irrelevant and merging never happens:
/sys/block/pmem0/queue/nomerges:0
/sys/block/pmem0/queue/nr_requests:128
Consider making them both read-only with:
* nomerges set to 2 (no merging happening)
* nr_requests as small as the block layer allows to avoid
wasting memory.
5. No scatter-gather lists are created by the driver, so these
read-only fields are meaningless:
/sys/block/pmem0/queue/max_segments:128
/sys/block/pmem0/queue/max_segment_size:65536
Is there a better way to report them as irrelevant?
6. There is no completion processing, so the read-writeable
cpu affinity is not used:
/sys/block/pmem0/queue/rq_affinity:0
Consider making it read-only and set to 2, meaning the
completions always run on the requesting CPU.
7. With mmap() allowing less than logical block sized accesses
to the device, this could be considered misleading:
/sys/block/pmem0/queue/physical_block_size:512
Perhaps that needs to be 1 byte or a cacheline size (64 bytes
on x86) to indicate that direct partial logical block accesses
are possible. The btt driver could report 512 as one indication
it is different.
I wouldn't be surprised if smaller values than the logical block
size confused some software, though.
---
Robert Elliott, HP Server Storage
On Tue, Apr 28, 2015 at 2:24 PM, Elliott, Robert (Server Storage)
<[email protected]> wrote:
>> -----Original Message-----
>> From: Linux-nvdimm [mailto:[email protected]] On Behalf Of
>> Dan Williams
>> Sent: Tuesday, April 28, 2015 1:24 PM
>> To: [email protected]
>> Cc: Neil Brown; Dave Chinner; H. Peter Anvin; Christoph Hellwig; Rafael J.
>> Wysocki; Robert Moore; Ingo Molnar; [email protected]; Jens Axboe;
>> Borislav Petkov; Thomas Gleixner; Greg KH; [email protected];
>> Andy Lutomirski; Andrew Morton; Linus Torvalds
>> Subject: [Linux-nvdimm] [PATCH v2 00/20] libnd: non-volatile memory device
>> support
>>
>> Changes since v1 [1]: Incorporates feedback received prior to April 24.
>
> Here are some comments on the sysfs properties reported for a pmem device.
> They are based on v1, but I don't think v2 changes anything.
>
> 1. This confuses lsblk (part of util-linux):
> /sys/block/pmem0/device/type:4
>
> lsblk shows:
> NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
> pmem0 251:0 0 8G 0 worm
> pmem1 251:16 0 8G 0 worm
> pmem2 251:32 0 8G 0 worm
> pmem3 251:48 0 8G 0 worm
> pmem4 251:64 0 8G 0 worm
> pmem5 251:80 0 8G 0 worm
> pmem6 251:96 0 8G 0 worm
> pmem7 251:112 0 8G 0 worm
>
> lsblk's blkdev_scsi_type_to_name() considers 4 to mean
> SCSI_TYPE_WORM (write once read many ... used for certain optical
> and tape drives).
Why is lsblk assuming these are scsi devices? I'll need to go check that out.
> I'm not sure what nd and pmem are doing to result in that value.
That is their libnd specific device type number from
include/uapi/ndctl.h. 4 == ND_DEVICE_NAMESPACE_IO. lsblk has no
business interpreting this as something SCSI specific.
> 2. To avoid confusing software trying to detect fast storage vs.
> slow storage devices via sysfs, this value should be 0:
> /sys/block/pmem0/queue/rotational:1
>
> That can be done by adding this shortly after the blk_alloc_queue call:
> queue_flag_set_unlocked(QUEUE_FLAG_NONROT, pmem->pmem_queue);
Yeah, good catch.
> 3. Is there any reason to have a 512 KiB limit on the transfer
> length?
> /sys/block/pmem0/queue/max_hw_sectors_kb:512
>
> That is from:
> blk_queue_max_hw_sectors(pmem->pmem_queue, 1024);
I'd only change this from the default if performance testing showed it
made a non-trivial difference.
> 4. These are read-writeable, but IOs never reach a queue, so
> the queue size is irrelevant and merging never happens:
> /sys/block/pmem0/queue/nomerges:0
> /sys/block/pmem0/queue/nr_requests:128
>
> Consider making them both read-only with:
> * nomerges set to 2 (no merging happening)
> * nr_requests as small as the block layer allows to avoid
> wasting memory.
>
> 5. No scatter-gather lists are created by the driver, so these
> read-only fields are meaningless:
> /sys/block/pmem0/queue/max_segments:128
> /sys/block/pmem0/queue/max_segment_size:65536
>
> Is there a better way to report them as irrelevant?
Again it comes back to the question of whether these default settings
are actively harmful.
>
> 6. There is no completion processing, so the read-writeable
> cpu affinity is not used:
> /sys/block/pmem0/queue/rq_affinity:0
>
> Consider making it read-only and set to 2, meaning the
> completions always run on the requesting CPU.
There are no completions with pmem, the entire I/O path is
synchronous. Ideally, this attribute would disappear for a pmem
queue, not be set to 2.
> 7. With mmap() allowing less than logical block sized accesses
> to the device, this could be considered misleading:
> /sys/block/pmem0/queue/physical_block_size:512
I don't see how it is misleading. If you access it as a block device
the block size is 512. If the application is mmap() + DAX aware it
knows that the physical_block_size is being bypassed.
>
> Perhaps that needs to be 1 byte or a cacheline size (64 bytes
> on x86) to indicate that direct partial logical block accesses
> are possible.
No, because that breaks the definition of a block device. Through the
bdev interface it's always accessed a block at a time.
> The btt driver could report 512 as one indication
> it is different.
>
> I wouldn't be surprised if smaller values than the logical block
> size confused some software, though.
Precisely why we shouldn't go there with pmem.
On Tue, Apr 28, 2015 at 2:04 PM, Andy Lutomirski <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 11:25 AM, Dan Williams <[email protected]> wrote:
>> +config ND_E820
>> + tristate "E820: Support the E820-type-12 PMEM convention"
>> + depends on X86_PMEM_LEGACY
>> + default m if X86_PMEM_LEGACY
>> + select LIBND
>> + help
>> + Prior to ACPI 6 some platforms advertised peristent memory
>> + via type-12 e820 memory ranges. Create a libnd bus and
>> + attach an instance of the pmem driver to these ranges.
>> +
>
> How about something like:
>
> "This driver allows libnd to work with legacy, pre-ACPI 6 NVDIMMs.
> This enables such devices to be exposed as block devices using PMEM.
>
> The legacy NVDIMM interface is problematic. This driver will not work
> if you boot using UEFI, and some NVDIMMs and motherboards that work
> with this driver may require proprietary code in order to work
> reliably."
Perhaps not "problematic" but "requires a BIOS in Legacy mode"
It might also mention that if you use the kernel command line
memmap=nn!ss syntax it adds
a type 12 region to the e820 map and so you would want this support.
If you have a motherboard with UEFI support for NVDIMM's that would be
the recommended
configuration.
Phil P.
--
Philip Pokorny, RHCE
Chief Technology Officer
PENGUIN COMPUTING, Inc
http://www.penguincomputing.com
On Tue, Apr 28, 2015 at 2:06 PM, Andy Lutomirski <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 1:59 PM, Dan Williams <[email protected]> wrote:
>> On Tue, Apr 28, 2015 at 1:52 PM, Andy Lutomirski <[email protected]> wrote:
>>> On Tue, Apr 28, 2015 at 11:24 AM, Dan Williams <[email protected]> wrote:
>>>> Changes since v1 [1]: Incorporates feedback received prior to April 24.
>>>>
>>>> 1/ Ingo said [2]:
>>>>
>>>> "So why on earth is this whole concept and the naming itself
>>>> ('drivers/block/nd/' stands for 'NFIT Defined', apparently)
>>>> revolving around a specific 'firmware' mindset and revolving
>>>> around specific, weirdly named, overly complicated looking
>>>> firmware interfaces that come with their own new weird
>>>> glossary??"
>>>>
>>>> Indeed, we of course consulted the NFIT specification to determine
>>>> the shape of the sub-system, but then let its terms and data
>>>> structures permeate too deep into the implementation. That is fixed
>>>> now with all NFIT specifics factored out into acpi.c. The NFIT is no
>>>> longer required reading to review libnd. Only three concepts are
>>>> needed:
>>>>
>>>> i/ PMEM - contiguous memory range where cpu stores are
>>>> persistent once they are flushed through the memory
>>>> controller.
>>>>
>>>> ii/ BLK - mmio apertures (sliding windows) that can be
>>>> programmed to access an aperture's-worth of persistent
>>>> media at a time.
>>>>
>>>> iii/ DPA - "dimm-physical-address", address space local to a
>>>> dimm. A dimm may provide both PMEM-mode and BLK-mode
>>>> access to a range of DPA. libnd manages allocation of DPA
>>>> to either PMEM or BLK-namespaces to resolve this aliasing.
>>>
>>> Mostly for my understanding: is there a name for "address relative to
>>> the address lines on the DIMM"? That is, a DIMM that exposes 8 GB of
>>> apparent physical memory, possibly interleaved, broken up, or weirdly
>>> remapped by the memory controller, would still have addresses between
>>> 0 and 8 GB. Some of those might be PMEM windows, some might be MMIO,
>>> some might be BLK apertures, etc.
>>>
>>> IIUC "DPA" refers to actual addressable storage, not this type of address?
>>
>> No, DPA is exactly as you describe above. You can't directly access
>> it except through a PMEM mapping (possibly interleaved with DPA from
>> other DIMMs) or a BLK aperture (mmio window into DPA).
>
> So the thing I'm describing has no name, then? Oh, well.
What? The thing you are describing *is* DPA.
On Tue, Apr 28, 2015 at 2:10 PM, Andy Lutomirski <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 11:26 AM, Dan Williams <[email protected]> wrote:
>> From: Ross Zwisler <[email protected]>
>>
>> The libnd implementation handles allocating dimm address space (DPA)
>> between PMEM and BLK mode interfaces. After DPA has been allocated from
>> a BLK-region to a BLK-namespace the nd_blk driver attaches to handle I/O
>> as a struct bio based block device. Unlike PMEM, BLK is required to
>> handle platform specific details like mmio register formats and memory
>> controller interleave. For this reason the libnd generic nd_blk driver
>> calls back into the bus provider to carry out the I/O.
>>
>> This initial implementation handles the BLK interface defined by the
>> ACPI 6 NFIT [1] and the NVDIMM DSM Interface Example [2] composed from
>> DCR (dimm control region), BDW (block data window), IDT (interleave
>> descriptor) NFIT structures and the hardware register format.
>> [1]: http://www.uefi.org/sites/default/files/resources/ACPI_6.0.pdf
>> [2]: http://pmem.io/documents/NVDIMM_DSM_Interface_Example.pdf
>>
>> Cc: Andy Lutomirski <[email protected]>
>> Cc: Boaz Harrosh <[email protected]>
>> Cc: H. Peter Anvin <[email protected]>
>> Cc: Jens Axboe <[email protected]>
>> Cc: Ingo Molnar <[email protected]>
>> Cc: Christoph Hellwig <[email protected]>
>> Signed-off-by: Ross Zwisler <[email protected]>
>> Signed-off-by: Dan Williams <[email protected]>
>> ---
>> drivers/block/nd/Kconfig | 12 +
>> drivers/block/nd/Makefile | 3
>> drivers/block/nd/acpi.c | 422 +++++++++++++++++++++++++++++++++++--
>> drivers/block/nd/acpi_nfit.h | 47 ++++
>> drivers/block/nd/blk.c | 264 +++++++++++++++++++++++
>> drivers/block/nd/libnd.h | 11 +
>> drivers/block/nd/namespace_devs.c | 47 ++++
>> drivers/block/nd/nd-private.h | 3
>> drivers/block/nd/nd.h | 16 +
>> drivers/block/nd/region.c | 8 +
>> drivers/block/nd/region_devs.c | 65 +++++-
>> drivers/block/nd/test/nfit.c | 29 +++
>> drivers/block/nd/test/nfit_test.h | 2
>> 13 files changed, 891 insertions(+), 38 deletions(-)
>> create mode 100644 drivers/block/nd/blk.c
>>
>> diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
>> index 612bf2b14283..bac4290129fc 100644
>> --- a/drivers/block/nd/Kconfig
>> +++ b/drivers/block/nd/Kconfig
>> @@ -95,6 +95,18 @@ config BLK_DEV_PMEM
>>
>> Say Y if you want to use a NVDIMM described by ACPI, E820, etc...
>>
>> +config ND_BLK
>> + tristate "BLK: Block data window (aperture) device support"
>> + depends on LIBND
>> + default ND_ACPI
>> + help
>> + This driver performs I/O using a set of mmio windows on a
>> + dimm. The set of apertures will all access the one DIMM.
>> + Multiple windows allow multiple threads to have a different
>> + portions of the dimm open at one time.
>> +
>> + Say Y if you want to use a NVDIMM with BLK-mode capability
>> +
>
> This describes how it works, not what it is. How about:
>
> This driver exposes NVDIMM BLK regions as block devices. BLK regions
> are regions of NVDIMM storage that are sector-addressable, not
> byte-addressible, and do not support DAX.
They *are* byte-addressable albeit through an indirection window. The
indirection windows are too small for DAX to be a viable access mode.
On Tue, Apr 28, 2015 at 3:21 PM, Phil Pokorny
<[email protected]> wrote:
> On Tue, Apr 28, 2015 at 2:04 PM, Andy Lutomirski <[email protected]> wrote:
>> On Tue, Apr 28, 2015 at 11:25 AM, Dan Williams <[email protected]> wrote:
>>> +config ND_E820
>>> + tristate "E820: Support the E820-type-12 PMEM convention"
>>> + depends on X86_PMEM_LEGACY
>>> + default m if X86_PMEM_LEGACY
>>> + select LIBND
>>> + help
>>> + Prior to ACPI 6 some platforms advertised peristent memory
>>> + via type-12 e820 memory ranges. Create a libnd bus and
>>> + attach an instance of the pmem driver to these ranges.
>>> +
>>
>> How about something like:
>>
>> "This driver allows libnd to work with legacy, pre-ACPI 6 NVDIMMs.
>> This enables such devices to be exposed as block devices using PMEM.
>>
>> The legacy NVDIMM interface is problematic. This driver will not work
>> if you boot using UEFI, and some NVDIMMs and motherboards that work
>> with this driver may require proprietary code in order to work
>> reliably."
>
> Perhaps not "problematic" but "requires a BIOS in Legacy mode"
>
> It might also mention that if you use the kernel command line
> memmap=nn!ss syntax it adds
> a type 12 region to the e820 map and so you would want this support.
>
> If you have a motherboard with UEFI support for NVDIMM's that would be
> the recommended
> configuration.
This is such a mess that I think this driver should maybe flat-out
refuse to load in this type of configuration without some scary module
option. I have some NVDIMMs that report as type 12 but need two extra
out-of-tree drivers to work safely. First, they need i2c_imc or the
equivalent (I'll try to resubmit that soon). Second, they need secret
magic NDAed register poking. The latter is very problematic.
At the very least, I think we should discourage people who don't
really know what they're doing from using this driver without care.
--Andy
On Tue, Apr 28, 2015 at 3:28 PM, Dan Williams <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 2:06 PM, Andy Lutomirski <[email protected]> wrote:
>> On Tue, Apr 28, 2015 at 1:59 PM, Dan Williams <[email protected]> wrote:
>>> On Tue, Apr 28, 2015 at 1:52 PM, Andy Lutomirski <[email protected]> wrote:
>>>> On Tue, Apr 28, 2015 at 11:24 AM, Dan Williams <[email protected]> wrote:
>>>>> Changes since v1 [1]: Incorporates feedback received prior to April 24.
>>>>>
>>>>> 1/ Ingo said [2]:
>>>>>
>>>>> "So why on earth is this whole concept and the naming itself
>>>>> ('drivers/block/nd/' stands for 'NFIT Defined', apparently)
>>>>> revolving around a specific 'firmware' mindset and revolving
>>>>> around specific, weirdly named, overly complicated looking
>>>>> firmware interfaces that come with their own new weird
>>>>> glossary??"
>>>>>
>>>>> Indeed, we of course consulted the NFIT specification to determine
>>>>> the shape of the sub-system, but then let its terms and data
>>>>> structures permeate too deep into the implementation. That is fixed
>>>>> now with all NFIT specifics factored out into acpi.c. The NFIT is no
>>>>> longer required reading to review libnd. Only three concepts are
>>>>> needed:
>>>>>
>>>>> i/ PMEM - contiguous memory range where cpu stores are
>>>>> persistent once they are flushed through the memory
>>>>> controller.
>>>>>
>>>>> ii/ BLK - mmio apertures (sliding windows) that can be
>>>>> programmed to access an aperture's-worth of persistent
>>>>> media at a time.
>>>>>
>>>>> iii/ DPA - "dimm-physical-address", address space local to a
>>>>> dimm. A dimm may provide both PMEM-mode and BLK-mode
>>>>> access to a range of DPA. libnd manages allocation of DPA
>>>>> to either PMEM or BLK-namespaces to resolve this aliasing.
>>>>
>>>> Mostly for my understanding: is there a name for "address relative to
>>>> the address lines on the DIMM"? That is, a DIMM that exposes 8 GB of
>>>> apparent physical memory, possibly interleaved, broken up, or weirdly
>>>> remapped by the memory controller, would still have addresses between
>>>> 0 and 8 GB. Some of those might be PMEM windows, some might be MMIO,
>>>> some might be BLK apertures, etc.
>>>>
>>>> IIUC "DPA" refers to actual addressable storage, not this type of address?
>>>
>>> No, DPA is exactly as you describe above. You can't directly access
>>> it except through a PMEM mapping (possibly interleaved with DPA from
>>> other DIMMs) or a BLK aperture (mmio window into DPA).
>>
>> So the thing I'm describing has no name, then? Oh, well.
>
> What? The thing you are describing *is* DPA.
I'm confused. Here are the two things I have in mind:
1. An address into on-DIMM storage. If I have a DIMM that is mapped
to 8 GB of SPA but has 64 GB of usable storage (accessed through BLK
apertures, say), then this address runs from 0 to 64 GB.
2. An address into the DIMM's view of physical address space. If I
have a DIMM that is mapped to 8 GB of SPA but has 64 GB of usable
storage (accessed through BLK apertures, say), then this address runs
from 0 to 8 GB. There's a one-to-one mapping between SPA and this
type of address.
Since you said "a dimm may provide both PMEM-mode and BLK-mode access
to a range of DPA.," I thought that DPA was #1.
--Andy
On Tue, Apr 28, 2015 at 3:30 PM, Dan Williams <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 2:10 PM, Andy Lutomirski <[email protected]> wrote:
>> On Tue, Apr 28, 2015 at 11:26 AM, Dan Williams <[email protected]> wrote:
>>> From: Ross Zwisler <[email protected]>
>>>
>>> The libnd implementation handles allocating dimm address space (DPA)
>>> between PMEM and BLK mode interfaces. After DPA has been allocated from
>>> a BLK-region to a BLK-namespace the nd_blk driver attaches to handle I/O
>>> as a struct bio based block device. Unlike PMEM, BLK is required to
>>> handle platform specific details like mmio register formats and memory
>>> controller interleave. For this reason the libnd generic nd_blk driver
>>> calls back into the bus provider to carry out the I/O.
>>>
>>> This initial implementation handles the BLK interface defined by the
>>> ACPI 6 NFIT [1] and the NVDIMM DSM Interface Example [2] composed from
>>> DCR (dimm control region), BDW (block data window), IDT (interleave
>>> descriptor) NFIT structures and the hardware register format.
>>> [1]: http://www.uefi.org/sites/default/files/resources/ACPI_6.0.pdf
>>> [2]: http://pmem.io/documents/NVDIMM_DSM_Interface_Example.pdf
>>>
>>> Cc: Andy Lutomirski <[email protected]>
>>> Cc: Boaz Harrosh <[email protected]>
>>> Cc: H. Peter Anvin <[email protected]>
>>> Cc: Jens Axboe <[email protected]>
>>> Cc: Ingo Molnar <[email protected]>
>>> Cc: Christoph Hellwig <[email protected]>
>>> Signed-off-by: Ross Zwisler <[email protected]>
>>> Signed-off-by: Dan Williams <[email protected]>
>>> ---
>>> drivers/block/nd/Kconfig | 12 +
>>> drivers/block/nd/Makefile | 3
>>> drivers/block/nd/acpi.c | 422 +++++++++++++++++++++++++++++++++++--
>>> drivers/block/nd/acpi_nfit.h | 47 ++++
>>> drivers/block/nd/blk.c | 264 +++++++++++++++++++++++
>>> drivers/block/nd/libnd.h | 11 +
>>> drivers/block/nd/namespace_devs.c | 47 ++++
>>> drivers/block/nd/nd-private.h | 3
>>> drivers/block/nd/nd.h | 16 +
>>> drivers/block/nd/region.c | 8 +
>>> drivers/block/nd/region_devs.c | 65 +++++-
>>> drivers/block/nd/test/nfit.c | 29 +++
>>> drivers/block/nd/test/nfit_test.h | 2
>>> 13 files changed, 891 insertions(+), 38 deletions(-)
>>> create mode 100644 drivers/block/nd/blk.c
>>>
>>> diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
>>> index 612bf2b14283..bac4290129fc 100644
>>> --- a/drivers/block/nd/Kconfig
>>> +++ b/drivers/block/nd/Kconfig
>>> @@ -95,6 +95,18 @@ config BLK_DEV_PMEM
>>>
>>> Say Y if you want to use a NVDIMM described by ACPI, E820, etc...
>>>
>>> +config ND_BLK
>>> + tristate "BLK: Block data window (aperture) device support"
>>> + depends on LIBND
>>> + default ND_ACPI
>>> + help
>>> + This driver performs I/O using a set of mmio windows on a
>>> + dimm. The set of apertures will all access the one DIMM.
>>> + Multiple windows allow multiple threads to have a different
>>> + portions of the dimm open at one time.
>>> +
>>> + Say Y if you want to use a NVDIMM with BLK-mode capability
>>> +
>>
>> This describes how it works, not what it is. How about:
>>
>> This driver exposes NVDIMM BLK regions as block devices. BLK regions
>> are regions of NVDIMM storage that are sector-addressable, not
>> byte-addressible, and do not support DAX.
>
> They *are* byte-addressable albeit through an indirection window. The
> indirection windows are too small for DAX to be a viable access mode.
Right, I was assuming incorrectly that the sector-atomic thing was a
necessary part of BLK, or at least of this implementation.
Anyway, I think my point stands: let's describe what these drivers do
from a user's perspective, not how they work.
--Andy
--
Andy Lutomirski
AMA Capital Management, LLC
On Tue, Apr 28, 2015 at 3:58 PM, Andy Lutomirski <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 3:21 PM, Phil Pokorny
> <[email protected]> wrote:
>> On Tue, Apr 28, 2015 at 2:04 PM, Andy Lutomirski <[email protected]> wrote:
>>> On Tue, Apr 28, 2015 at 11:25 AM, Dan Williams <[email protected]> wrote:
>>>> +config ND_E820
>>>> + tristate "E820: Support the E820-type-12 PMEM convention"
>>>> + depends on X86_PMEM_LEGACY
>>>> + default m if X86_PMEM_LEGACY
>>>> + select LIBND
>>>> + help
>>>> + Prior to ACPI 6 some platforms advertised peristent memory
>>>> + via type-12 e820 memory ranges. Create a libnd bus and
>>>> + attach an instance of the pmem driver to these ranges.
>>>> +
>>>
>>> How about something like:
>>>
>>> "This driver allows libnd to work with legacy, pre-ACPI 6 NVDIMMs.
>>> This enables such devices to be exposed as block devices using PMEM.
>>>
>>> The legacy NVDIMM interface is problematic. This driver will not work
>>> if you boot using UEFI, and some NVDIMMs and motherboards that work
>>> with this driver may require proprietary code in order to work
>>> reliably."
>>
>> Perhaps not "problematic" but "requires a BIOS in Legacy mode"
>>
>> It might also mention that if you use the kernel command line
>> memmap=nn!ss syntax it adds
>> a type 12 region to the e820 map and so you would want this support.
>>
>> If you have a motherboard with UEFI support for NVDIMM's that would be
>> the recommended
>> configuration.
>
> This is such a mess that I think this driver should maybe flat-out
> refuse to load in this type of configuration without some scary module
> option. I have some NVDIMMs that report as type 12 but need two extra
> out-of-tree drivers to work safely. First, they need i2c_imc or the
> equivalent (I'll try to resubmit that soon). Second, they need secret
> magic NDAed register poking. The latter is very problematic.
My current experience is that things may be changing to something of a de-facto
standard in the area of register poking. In which case, we should be
able to ask
the de-facto vendor standard to be released under a non-NDA license so we can
write a proper user-space library for it. Or at worst, get a
proprietary source utility
that can do the poking.
The vendor isn't going to sell anything if they don't provide the
tools their resellers
and customers need.
> At the very least, I think we should discourage people who don't
> really know what they're doing from using this driver without care.
What would be the fun in that...
But seriously, speaking as Penguin Computing and a retailer of
hardware, I'd rather
not have the kernel telling my customers what's safe and what isn't
when it's a matter
of opinion. We provide a solution with support and having to tell my
customers: "you
need to load the module with the 'THIS_IS_UNSAFE' argument set to 3"
isn't productive.
Another intersesting possibility of the memmap= directive to declare a
type 12 region of
of memory is that you can test the driver (without the persistance) on
any arbitrary region
of memory in a machine. Other comments on this patch set talked about
having to put
virtual test hardware in qemu or kvm. Aside from the register poking,
just adding
memmap=xx!yy to the command line gives you something pmem can attach to and
you can use to test with. I suppose you could even simulate
persistance by saving off
the contents and restoring it on a controlled reboot.
Phil P.
--
Philip Pokorny, RHCE
Chief Technology Officer
PENGUIN COMPUTING, Inc
http://www.penguincomputing.com
On Tuesday, April 28, 2015 02:24:12 PM Dan Williams wrote:
> Changes since v1 [1]: Incorporates feedback received prior to April 24.
>
> 1/ Ingo said [2]:
>
> "So why on earth is this whole concept and the naming itself
> ('drivers/block/nd/' stands for 'NFIT Defined', apparently)
> revolving around a specific 'firmware' mindset and revolving
> around specific, weirdly named, overly complicated looking
> firmware interfaces that come with their own new weird
> glossary??"
>
> Indeed, we of course consulted the NFIT specification to determine
> the shape of the sub-system, but then let its terms and data
> structures permeate too deep into the implementation. That is fixed
> now with all NFIT specifics factored out into acpi.c. The NFIT is no
> longer required reading to review libnd. Only three concepts are
> needed:
>
> i/ PMEM - contiguous memory range where cpu stores are
> persistent once they are flushed through the memory
> controller.
>
> ii/ BLK - mmio apertures (sliding windows) that can be
> programmed to access an aperture's-worth of persistent
> media at a time.
>
> iii/ DPA - "dimm-physical-address", address space local to a
> dimm. A dimm may provide both PMEM-mode and BLK-mode
> access to a range of DPA. libnd manages allocation of DPA
> to either PMEM or BLK-namespaces to resolve this aliasing.
>
> The v1..v2 diffstat below shows the migration of nfit-specifics to
> acpi.c and the new state of libnd being nfit-free. "nd" now only
> refers to "non-volatile devices". Note, reworked documentation will
> return once the review has settled.
>
> Documentation/blockdev/nd.txt | 867 ---------------------
> MAINTAINERS | 34 +-
> arch/ia64/kernel/efi.c | 5 +-
> arch/x86/kernel/e820.c | 11 +-
> arch/x86/kernel/pmem.c | 2 +-
> drivers/block/Makefile | 2 +-
> drivers/block/nd/Kconfig | 135 ++--
> drivers/block/nd/Makefile | 32 +-
> drivers/block/nd/acpi.c | 1506 +++++++++++++++++++++++++++++++------
> drivers/block/nd/acpi_nfit.h | 321 ++++++++
> drivers/block/nd/blk.c | 27 +-
> drivers/block/nd/btt.c | 6 +-
> drivers/block/nd/btt_devs.c | 8 +-
> drivers/block/nd/bus.c | 337 +++++----
> drivers/block/nd/core.c | 574 +-------------
> drivers/block/nd/dimm.c | 11 -
> drivers/block/nd/dimm_devs.c | 292 ++-----
> drivers/block/nd/e820.c | 100 +++
> drivers/block/nd/libnd.h | 122 +++
> drivers/block/nd/namespace_devs.c | 10 +-
> drivers/block/nd/nd-private.h | 107 +--
> drivers/block/nd/nd.h | 91 +--
> drivers/block/nd/nfit.h | 238 ------
> drivers/block/nd/pmem.c | 56 +-
> drivers/block/nd/region.c | 78 +-
> drivers/block/nd/region_devs.c | 783 +++----------------
> drivers/block/nd/test/iomap.c | 86 +--
> drivers/block/nd/test/nfit.c | 1115 +++++++++++++++------------
> drivers/block/nd/test/nfit_test.h | 15 +-
> include/uapi/linux/ndctl.h | 130 ++--
> 30 files changed, 3166 insertions(+), 3935 deletions(-)
> delete mode 100644 Documentation/blockdev/nd.txt
> create mode 100644 drivers/block/nd/acpi_nfit.h
> create mode 100644 drivers/block/nd/e820.c
> create mode 100644 drivers/block/nd/libnd.h
> delete mode 100644 drivers/block/nd/nfit.h
>
> [1]: https://lists.01.org/pipermail/linux-nvdimm/2015-April/000484.html
> [2]: https://lists.01.org/pipermail/linux-nvdimm/2015-April/000520.html
>
> 2/ Christoph asked the pmem ida conversion to be moved to its own patch
> (done), and to consider leaving the current pmem.c in drivers/block/.
> Instead, I converted the e820-type-12 enabling to be the first
> non-ACPI-NFIT based consumer of libnd. The new nd_e820 driver simply
> registers e820-type-12 ranges as libnd PMEM regions. Among other
> things this conversion enables BTT for these ranges. The alternative
> is to move drivers/block/nd/nd.h internals out to include/linux/
> which I think is worse.
>
> 3/ Toshi reported that the NFIT parsing fails to handle the case of a
> PMEM range with a single-dimm (non-aliasing) interleave description.
> Support for this case was added and is tested by default by the
> nfit_test.1 configuration.
>
> 4/ Toshi reported that we should not be treating a missing _STA property
> as a "dimm disabled by firmware" case. (fixed).
>
> 5/ Christoph noted that ND_ARCH_HAS_IOREMAP_CACHE needs to be moved to
> arch code. It is gone for now and we'll revisit when adding cached
> mappings back to the PMEM driver.
>
> 6/ Toshi mentioned that the presence of two different nd_bus_probe()
> functions was confusing. (cleaned up).
>
> 7/ Robert asked for s/btt_checksum/nd_btt_checksum/ (done).
>
> 8/ Linda asked for nfit_test to honor dynamic cma reservations via the
> cma= command line (done). The cma requirements have also been
> reduced to 128M as only the simulated DAX regions need CMA. The rest
> can use vmalloc().
>
> ---
>
> Available here:
> git://git.kernel.org/pub/scm/linux/kernel/git/djbw/nvdimm nd-v2
>
> ---
>
> Dan Williams (18):
> e820, efi: add ACPI 6.0 persistent memory types
> libnd, nd_acpi: initial libnd infrastructure and NFIT support
> nd_acpi, nfit-test: manufactured NFITs for interface development
> libnd: ndctl class device, and nd bus attributes
> libnd, nd_acpi: dimm/memory-devices
> libnd: ndctl.h, the nd ioctl abi
> libnd, nd_dimm: dimm driver and base libnd device-driver infrastructure
> libnd, nd_acpi: regions (block-data-window, persistent memory, volatile memory)
> libnd: support for legacy (non-aliasing) nvdimms
> pmem: use ida
> libnd, nd_pmem: add libnd support to the pmem driver
> libnd, nd_acpi: add interleave-set state-tracking infrastructure
> libnd: namespace indices: read and validate
> libnd: pmem label sets and namespace instantiation.
> libnd: blk labels and namespace instantiation
> libnd: write pmem label set
> libnd: write blk label set
> libnd: infrastructure for btt devices
>
> Ross Zwisler (1):
> libnd, nd_acpi, nd_blk: driver for BLK-mode access persistent memory
>
> Vishal Verma (1):
> nd_btt: atomic sector updates
I'm wondering what's wrong with CCing all of the series to linux-acpi?
Is there anything in it that the people on that list should not see, by any
chance?
Like patch [01/20], for the most obvious example. It even has "ACPI 6" in the
subject ...
Rafael
On Tue, Apr 28, 2015 at 5:17 PM, Phil Pokorny
<[email protected]> wrote:
> On Tue, Apr 28, 2015 at 3:58 PM, Andy Lutomirski <[email protected]> wrote:
>> On Tue, Apr 28, 2015 at 3:21 PM, Phil Pokorny
>> <[email protected]> wrote:
>>> On Tue, Apr 28, 2015 at 2:04 PM, Andy Lutomirski <[email protected]> wrote:
>>>> On Tue, Apr 28, 2015 at 11:25 AM, Dan Williams <[email protected]> wrote:
>>>>> +config ND_E820
>>>>> + tristate "E820: Support the E820-type-12 PMEM convention"
>>>>> + depends on X86_PMEM_LEGACY
>>>>> + default m if X86_PMEM_LEGACY
>>>>> + select LIBND
>>>>> + help
>>>>> + Prior to ACPI 6 some platforms advertised peristent memory
>>>>> + via type-12 e820 memory ranges. Create a libnd bus and
>>>>> + attach an instance of the pmem driver to these ranges.
>>>>> +
>>>>
>>>> How about something like:
>>>>
>>>> "This driver allows libnd to work with legacy, pre-ACPI 6 NVDIMMs.
>>>> This enables such devices to be exposed as block devices using PMEM.
>>>>
>>>> The legacy NVDIMM interface is problematic. This driver will not work
>>>> if you boot using UEFI, and some NVDIMMs and motherboards that work
>>>> with this driver may require proprietary code in order to work
>>>> reliably."
>>>
>>> Perhaps not "problematic" but "requires a BIOS in Legacy mode"
>>>
>>> It might also mention that if you use the kernel command line
>>> memmap=nn!ss syntax it adds
>>> a type 12 region to the e820 map and so you would want this support.
>>>
>>> If you have a motherboard with UEFI support for NVDIMM's that would be
>>> the recommended
>>> configuration.
>>
>> This is such a mess that I think this driver should maybe flat-out
>> refuse to load in this type of configuration without some scary module
>> option. I have some NVDIMMs that report as type 12 but need two extra
>> out-of-tree drivers to work safely. First, they need i2c_imc or the
>> equivalent (I'll try to resubmit that soon). Second, they need secret
>> magic NDAed register poking. The latter is very problematic.
>
> My current experience is that things may be changing to something of a de-facto
> standard in the area of register poking. In which case, we should be
> able to ask
> the de-facto vendor standard to be released under a non-NDA license so we can
> write a proper user-space library for it. Or at worst, get a
> proprietary source utility
> that can do the poking.
>
> The vendor isn't going to sell anything if they don't provide the
> tools their resellers
> and customers need.
>
I suspect that the vendor will soon be done selling this particular
part as they move toward something more standard. Dunno.
>
>> At the very least, I think we should discourage people who don't
>> really know what they're doing from using this driver without care.
>
> What would be the fun in that...
>
> But seriously, speaking as Penguin Computing and a retailer of
> hardware, I'd rather
> not have the kernel telling my customers what's safe and what isn't
> when it's a matter
> of opinion. We provide a solution with support and having to tell my
> customers: "you
> need to load the module with the 'THIS_IS_UNSAFE' argument set to 3"
> isn't productive.
It could be that you load with i_promise_i_have_an_nvdimm_driver_too=1
or, better yet, if loaded without the magic option but with the magic
driver it figures it out and initializes anyway.
>
> Another intersesting possibility of the memmap= directive to declare a
> type 12 region of
> of memory is that you can test the driver (without the persistance) on
> any arbitrary region
> of memory in a machine. Other comments on this patch set talked about
> having to put
> virtual test hardware in qemu or kvm. Aside from the register poking,
> just adding
> memmap=xx!yy to the command line gives you something pmem can attach to and
> you can use to test with. I suppose you could even simulate
> persistance by saving off
> the contents and restoring it on a controlled reboot.
That's definitely useful.
Anyway, I don't object strongly to the driver as is. Anyone with a
legacy NVDIMM is already dependent on all kinds of things going right
(correct power supply, ADR, all pins wired correctly, correct BIOS
version, no EFI, lack of pcommit not being a problem, etc).
--Andy
On Tue, Apr 28, 2015 at 5:25 PM, Rafael J. Wysocki <[email protected]> wrote:
> On Tuesday, April 28, 2015 02:24:12 PM Dan Williams wrote:
>> Changes since v1 [1]: Incorporates feedback received prior to April 24.
>>
>> 1/ Ingo said [2]:
>>
>> "So why on earth is this whole concept and the naming itself
>> ('drivers/block/nd/' stands for 'NFIT Defined', apparently)
>> revolving around a specific 'firmware' mindset and revolving
>> around specific, weirdly named, overly complicated looking
>> firmware interfaces that come with their own new weird
>> glossary??"
>>
>> Indeed, we of course consulted the NFIT specification to determine
>> the shape of the sub-system, but then let its terms and data
>> structures permeate too deep into the implementation. That is fixed
>> now with all NFIT specifics factored out into acpi.c. The NFIT is no
>> longer required reading to review libnd. Only three concepts are
>> needed:
>>
>> i/ PMEM - contiguous memory range where cpu stores are
>> persistent once they are flushed through the memory
>> controller.
>>
>> ii/ BLK - mmio apertures (sliding windows) that can be
>> programmed to access an aperture's-worth of persistent
>> media at a time.
>>
>> iii/ DPA - "dimm-physical-address", address space local to a
>> dimm. A dimm may provide both PMEM-mode and BLK-mode
>> access to a range of DPA. libnd manages allocation of DPA
>> to either PMEM or BLK-namespaces to resolve this aliasing.
>>
>> The v1..v2 diffstat below shows the migration of nfit-specifics to
>> acpi.c and the new state of libnd being nfit-free. "nd" now only
>> refers to "non-volatile devices". Note, reworked documentation will
>> return once the review has settled.
>>
>> Documentation/blockdev/nd.txt | 867 ---------------------
>> MAINTAINERS | 34 +-
>> arch/ia64/kernel/efi.c | 5 +-
>> arch/x86/kernel/e820.c | 11 +-
>> arch/x86/kernel/pmem.c | 2 +-
>> drivers/block/Makefile | 2 +-
>> drivers/block/nd/Kconfig | 135 ++--
>> drivers/block/nd/Makefile | 32 +-
>> drivers/block/nd/acpi.c | 1506 +++++++++++++++++++++++++++++++------
>> drivers/block/nd/acpi_nfit.h | 321 ++++++++
>> drivers/block/nd/blk.c | 27 +-
>> drivers/block/nd/btt.c | 6 +-
>> drivers/block/nd/btt_devs.c | 8 +-
>> drivers/block/nd/bus.c | 337 +++++----
>> drivers/block/nd/core.c | 574 +-------------
>> drivers/block/nd/dimm.c | 11 -
>> drivers/block/nd/dimm_devs.c | 292 ++-----
>> drivers/block/nd/e820.c | 100 +++
>> drivers/block/nd/libnd.h | 122 +++
>> drivers/block/nd/namespace_devs.c | 10 +-
>> drivers/block/nd/nd-private.h | 107 +--
>> drivers/block/nd/nd.h | 91 +--
>> drivers/block/nd/nfit.h | 238 ------
>> drivers/block/nd/pmem.c | 56 +-
>> drivers/block/nd/region.c | 78 +-
>> drivers/block/nd/region_devs.c | 783 +++----------------
>> drivers/block/nd/test/iomap.c | 86 +--
>> drivers/block/nd/test/nfit.c | 1115 +++++++++++++++------------
>> drivers/block/nd/test/nfit_test.h | 15 +-
>> include/uapi/linux/ndctl.h | 130 ++--
>> 30 files changed, 3166 insertions(+), 3935 deletions(-)
>> delete mode 100644 Documentation/blockdev/nd.txt
>> create mode 100644 drivers/block/nd/acpi_nfit.h
>> create mode 100644 drivers/block/nd/e820.c
>> create mode 100644 drivers/block/nd/libnd.h
>> delete mode 100644 drivers/block/nd/nfit.h
>>
>> [1]: https://lists.01.org/pipermail/linux-nvdimm/2015-April/000484.html
>> [2]: https://lists.01.org/pipermail/linux-nvdimm/2015-April/000520.html
>>
>> 2/ Christoph asked the pmem ida conversion to be moved to its own patch
>> (done), and to consider leaving the current pmem.c in drivers/block/.
>> Instead, I converted the e820-type-12 enabling to be the first
>> non-ACPI-NFIT based consumer of libnd. The new nd_e820 driver simply
>> registers e820-type-12 ranges as libnd PMEM regions. Among other
>> things this conversion enables BTT for these ranges. The alternative
>> is to move drivers/block/nd/nd.h internals out to include/linux/
>> which I think is worse.
>>
>> 3/ Toshi reported that the NFIT parsing fails to handle the case of a
>> PMEM range with a single-dimm (non-aliasing) interleave description.
>> Support for this case was added and is tested by default by the
>> nfit_test.1 configuration.
>>
>> 4/ Toshi reported that we should not be treating a missing _STA property
>> as a "dimm disabled by firmware" case. (fixed).
>>
>> 5/ Christoph noted that ND_ARCH_HAS_IOREMAP_CACHE needs to be moved to
>> arch code. It is gone for now and we'll revisit when adding cached
>> mappings back to the PMEM driver.
>>
>> 6/ Toshi mentioned that the presence of two different nd_bus_probe()
>> functions was confusing. (cleaned up).
>>
>> 7/ Robert asked for s/btt_checksum/nd_btt_checksum/ (done).
>>
>> 8/ Linda asked for nfit_test to honor dynamic cma reservations via the
>> cma= command line (done). The cma requirements have also been
>> reduced to 128M as only the simulated DAX regions need CMA. The rest
>> can use vmalloc().
>>
>> ---
>>
>> Available here:
>> git://git.kernel.org/pub/scm/linux/kernel/git/djbw/nvdimm nd-v2
>>
>> ---
>>
>> Dan Williams (18):
>> e820, efi: add ACPI 6.0 persistent memory types
>> libnd, nd_acpi: initial libnd infrastructure and NFIT support
>> nd_acpi, nfit-test: manufactured NFITs for interface development
>> libnd: ndctl class device, and nd bus attributes
>> libnd, nd_acpi: dimm/memory-devices
>> libnd: ndctl.h, the nd ioctl abi
>> libnd, nd_dimm: dimm driver and base libnd device-driver infrastructure
>> libnd, nd_acpi: regions (block-data-window, persistent memory, volatile memory)
>> libnd: support for legacy (non-aliasing) nvdimms
>> pmem: use ida
>> libnd, nd_pmem: add libnd support to the pmem driver
>> libnd, nd_acpi: add interleave-set state-tracking infrastructure
>> libnd: namespace indices: read and validate
>> libnd: pmem label sets and namespace instantiation.
>> libnd: blk labels and namespace instantiation
>> libnd: write pmem label set
>> libnd: write blk label set
>> libnd: infrastructure for btt devices
>>
>> Ross Zwisler (1):
>> libnd, nd_acpi, nd_blk: driver for BLK-mode access persistent memory
>>
>> Vishal Verma (1):
>> nd_btt: atomic sector updates
>
> I'm wondering what's wrong with CCing all of the series to linux-acpi?
>
> Is there anything in it that the people on that list should not see, by any
> chance?
linux-acpi may not care about the dimm-metadata labeling patches that
are completely independent of ACPI, but might as well include
linux-acpi on the whole series at this point.
> -----Original Message-----
> From: Linux-nvdimm [mailto:[email protected]] On Behalf Of
> Dan Williams
> Sent: Tuesday, April 28, 2015 1:25 PM
> Subject: [Linux-nvdimm] [PATCH v2 08/20] libnd, nd_acpi: regions (block-
> data-window, persistent memory, volatile memory)
>
> A "region" device represents the maximum capacity of a BLK range (mmio
> block-data-window(s)), or a PMEM range (DAX-capable persistent memory or
> volatile memory), without regard for aliasing. Aliasing, in the
> dimm-local address space (DPA), is resolved by metadata on a dimm to
> designate which exclusive interface will access the aliased DPA ranges.
> Support for the per-dimm metadata/label arrvies is in a subsequent
> patch.
>
> The name format of "region" devices is "regionN" where, like dimms, N is
> a global ida index assigned at discovery time. This id is not reliable
> across reboots nor in the presence of hotplug. Look to attributes of
> the region or static id-data of the sub-namespace to generate a
> persistent name.
...
> +++ b/drivers/block/nd/region_devs.c
...
> +static noinline struct nd_region *nd_region_create(struct nd_bus *nd_bus,
> + struct nd_region_desc *ndr_desc, struct device_type *dev_type)
> +{
> + struct nd_region *nd_region;
> + struct device *dev;
> + u16 i;
> +
> + for (i = 0; i < ndr_desc->num_mappings; i++) {
> + struct nd_mapping *nd_mapping = &ndr_desc->nd_mapping[i];
> + struct nd_dimm *nd_dimm = nd_mapping->nd_dimm;
> +
> + if ((nd_mapping->start | nd_mapping->size) % SZ_4K) {
> + dev_err(&nd_bus->dev, "%pf: %s mapping%d is not 4K
> aligned\n",
> + __builtin_return_address(0),
Please use "KiB" rather than the unclear "K".
Same comment for a dev_dbg print in patch 14.
> + dev_name(&nd_dimm->dev), i);
> +
> + return NULL;
> + }
> + }
> +
> + nd_region = kzalloc(sizeof(struct nd_region)
> + + sizeof(struct nd_mapping) * ndr_desc->num_mappings,
> + GFP_KERNEL);
> + if (!nd_region)
> + return NULL;
> + nd_region->id = ida_simple_get(®ion_ida, 0, 0, GFP_KERNEL);
> + if (nd_region->id < 0) {
> + kfree(nd_region);
> + return NULL;
> + }
> +
> + memcpy(nd_region->mapping, ndr_desc->nd_mapping,
> + sizeof(struct nd_mapping) * ndr_desc->num_mappings);
> + for (i = 0; i < ndr_desc->num_mappings; i++) {
> + struct nd_mapping *nd_mapping = &ndr_desc->nd_mapping[i];
> + struct nd_dimm *nd_dimm = nd_mapping->nd_dimm;
> +
> + get_device(&nd_dimm->dev);
> + }
> + nd_region->ndr_mappings = ndr_desc->num_mappings;
> + nd_region->provider_data = ndr_desc->provider_data;
> + dev = &nd_region->dev;
> + dev_set_name(dev, "region%d", nd_region->id);
Could this include "nd" in the name, like "ndregion%d"?
The other dev_set_name calls in this patch set use:
btt%d
ndbus%d
nmem%d
namespace%d.%d
which are a bit more distinctive.
---
Robert Elliott, HP Server Storage
On Tue, Apr 28, 2015 at 3:58 PM, Andy Lutomirski <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 3:21 PM, Phil Pokorny
> <[email protected]> wrote:
>> On Tue, Apr 28, 2015 at 2:04 PM, Andy Lutomirski <[email protected]> wrote:
>>> On Tue, Apr 28, 2015 at 11:25 AM, Dan Williams <[email protected]> wrote:
[..]
> This is such a mess that I think this driver should maybe flat-out
> refuse to load in this type of configuration without some scary module
> option. I have some NVDIMMs that report as type 12 but need two extra
> out-of-tree drivers to work safely. First, they need i2c_imc or the
> equivalent (I'll try to resubmit that soon). Second, they need secret
> magic NDAed register poking. The latter is very problematic.
>
> At the very least, I think we should discourage people who don't
> really know what they're doing from using this driver without care.
The benefit of the type-12 experiment having not made it very far out
of the lab is that it may be feasible to whitelist known platforms
where we believe ADR is available. Otherwise, the presence of the
NFIT asserts platform persistent memory support.
On Wed, Apr 29, 2015 at 8:53 AM, Elliott, Robert (Server Storage)
<[email protected]> wrote:
>> -----Original Message-----
>> From: Linux-nvdimm [mailto:[email protected]] On Behalf Of
>> Dan Williams
>> Sent: Tuesday, April 28, 2015 1:25 PM
>> Subject: [Linux-nvdimm] [PATCH v2 08/20] libnd, nd_acpi: regions (block-
>> data-window, persistent memory, volatile memory)
>>
>> A "region" device represents the maximum capacity of a BLK range (mmio
>> block-data-window(s)), or a PMEM range (DAX-capable persistent memory or
>> volatile memory), without regard for aliasing. Aliasing, in the
>> dimm-local address space (DPA), is resolved by metadata on a dimm to
>> designate which exclusive interface will access the aliased DPA ranges.
>> Support for the per-dimm metadata/label arrvies is in a subsequent
>> patch.
>>
>> The name format of "region" devices is "regionN" where, like dimms, N is
>> a global ida index assigned at discovery time. This id is not reliable
>> across reboots nor in the presence of hotplug. Look to attributes of
>> the region or static id-data of the sub-namespace to generate a
>> persistent name.
> ...
>> +++ b/drivers/block/nd/region_devs.c
> ...
>> +static noinline struct nd_region *nd_region_create(struct nd_bus *nd_bus,
>> + struct nd_region_desc *ndr_desc, struct device_type *dev_type)
>> +{
>> + struct nd_region *nd_region;
>> + struct device *dev;
>> + u16 i;
>> +
>> + for (i = 0; i < ndr_desc->num_mappings; i++) {
>> + struct nd_mapping *nd_mapping = &ndr_desc->nd_mapping[i];
>> + struct nd_dimm *nd_dimm = nd_mapping->nd_dimm;
>> +
>> + if ((nd_mapping->start | nd_mapping->size) % SZ_4K) {
>> + dev_err(&nd_bus->dev, "%pf: %s mapping%d is not 4K
>> aligned\n",
>> + __builtin_return_address(0),
>
> Please use "KiB" rather than the unclear "K".
Ok.
> Same comment for a dev_dbg print in patch 14.
It's a debug statement, but ok.
[..]
>
> Could this include "nd" in the name, like "ndregion%d"?
>
> The other dev_set_name calls in this patch set use:
> btt%d
> ndbus%d
> nmem%d
> namespace%d.%d
>
> which are a bit more distinctive.
They sit on an "nd" bus and don't have global device nodes, I don't
see a need to make them anymore distinctive.
On Tue, Apr 28, 2015 at 4:06 PM, Andy Lutomirski <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 3:30 PM, Dan Williams <[email protected]> wrote:
>> On Tue, Apr 28, 2015 at 2:10 PM, Andy Lutomirski <[email protected]> wrote:
>>> On Tue, Apr 28, 2015 at 11:26 AM, Dan Williams <[email protected]> wrote:
>>>> From: Ross Zwisler <[email protected]>
>>>>
>>>> The libnd implementation handles allocating dimm address space (DPA)
>>>> between PMEM and BLK mode interfaces. After DPA has been allocated from
>>>> a BLK-region to a BLK-namespace the nd_blk driver attaches to handle I/O
>>>> as a struct bio based block device. Unlike PMEM, BLK is required to
>>>> handle platform specific details like mmio register formats and memory
>>>> controller interleave. For this reason the libnd generic nd_blk driver
>>>> calls back into the bus provider to carry out the I/O.
>>>>
>>>> This initial implementation handles the BLK interface defined by the
>>>> ACPI 6 NFIT [1] and the NVDIMM DSM Interface Example [2] composed from
>>>> DCR (dimm control region), BDW (block data window), IDT (interleave
>>>> descriptor) NFIT structures and the hardware register format.
>>>> [1]: http://www.uefi.org/sites/default/files/resources/ACPI_6.0.pdf
>>>> [2]: http://pmem.io/documents/NVDIMM_DSM_Interface_Example.pdf
>>>>
>>>> Cc: Andy Lutomirski <[email protected]>
>>>> Cc: Boaz Harrosh <[email protected]>
>>>> Cc: H. Peter Anvin <[email protected]>
>>>> Cc: Jens Axboe <[email protected]>
>>>> Cc: Ingo Molnar <[email protected]>
>>>> Cc: Christoph Hellwig <[email protected]>
>>>> Signed-off-by: Ross Zwisler <[email protected]>
>>>> Signed-off-by: Dan Williams <[email protected]>
>>>> ---
>>>> drivers/block/nd/Kconfig | 12 +
>>>> drivers/block/nd/Makefile | 3
>>>> drivers/block/nd/acpi.c | 422 +++++++++++++++++++++++++++++++++++--
>>>> drivers/block/nd/acpi_nfit.h | 47 ++++
>>>> drivers/block/nd/blk.c | 264 +++++++++++++++++++++++
>>>> drivers/block/nd/libnd.h | 11 +
>>>> drivers/block/nd/namespace_devs.c | 47 ++++
>>>> drivers/block/nd/nd-private.h | 3
>>>> drivers/block/nd/nd.h | 16 +
>>>> drivers/block/nd/region.c | 8 +
>>>> drivers/block/nd/region_devs.c | 65 +++++-
>>>> drivers/block/nd/test/nfit.c | 29 +++
>>>> drivers/block/nd/test/nfit_test.h | 2
>>>> 13 files changed, 891 insertions(+), 38 deletions(-)
>>>> create mode 100644 drivers/block/nd/blk.c
>>>>
>>>> diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
>>>> index 612bf2b14283..bac4290129fc 100644
>>>> --- a/drivers/block/nd/Kconfig
>>>> +++ b/drivers/block/nd/Kconfig
>>>> @@ -95,6 +95,18 @@ config BLK_DEV_PMEM
>>>>
>>>> Say Y if you want to use a NVDIMM described by ACPI, E820, etc...
>>>>
>>>> +config ND_BLK
>>>> + tristate "BLK: Block data window (aperture) device support"
>>>> + depends on LIBND
>>>> + default ND_ACPI
>>>> + help
>>>> + This driver performs I/O using a set of mmio windows on a
>>>> + dimm. The set of apertures will all access the one DIMM.
>>>> + Multiple windows allow multiple threads to have a different
>>>> + portions of the dimm open at one time.
>>>> +
>>>> + Say Y if you want to use a NVDIMM with BLK-mode capability
>>>> +
>>>
>>> This describes how it works, not what it is. How about:
>>>
>>> This driver exposes NVDIMM BLK regions as block devices. BLK regions
>>> are regions of NVDIMM storage that are sector-addressable, not
>>> byte-addressible, and do not support DAX.
>>
>> They *are* byte-addressable albeit through an indirection window. The
>> indirection windows are too small for DAX to be a viable access mode.
>
> Right, I was assuming incorrectly that the sector-atomic thing was a
> necessary part of BLK, or at least of this implementation.
>
> Anyway, I think my point stands: let's describe what these drivers do
> from a user's perspective, not how they work.
Agreed. How about:
"Support NVDIMMs, or other devices, that implement a BLK-mode access
capability. BLK-mode access uses memory-mapped-i/o apertures to
access persistent media.
Say Y if your platform firmware emits an ACPI.NFIT table
(CONFIG_ND_ACPI), or otherwise exposes BLK-mode capabilities."
Hi Dan,
Thanks for the update. This version of the patchset enumerates our NFIT
table properly. :-)
On Tue, 2015-04-28 at 14:25 -0400, Dan Williams wrote:
> In preparation for the pmem driver attaching to pmem-namespaces emitted
> by libnd, convert it to use an ida instead of an always increasing
> atomic index. This provides a bit of stability to pmem device names in
> the presence of driver re-bind events.
:
> @@ -122,20 +123,26 @@ static struct pmem_device *pmem_alloc(struct device *dev, struct resource *res)
> {
> struct pmem_device *pmem;
> struct gendisk *disk;
> - int idx, err;
> + int err;
>
> err = -ENOMEM;
> pmem = kzalloc(sizeof(*pmem), GFP_KERNEL);
> if (!pmem)
> goto out;
>
> + pmem->id = ida_simple_get(&pmem_ida, 0, 0, GFP_KERNEL);
nd_pmem_probe() is called asynchronously via async_schedule_domain
(). We have seen a case that the region#->pmem# binding becomes
inconsistent across a reboot when there are 8 NVDIMM cards (reported by
Robert Elliott). This leads user to access a wrong device.
I think pmem id needs to be assigned before async_schedule_domain(), and
cascaded to nd_pmem_probe().
-Toshi
On Wed, Apr 29, 2015 at 8:55 AM, Dan Williams <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 3:58 PM, Andy Lutomirski <[email protected]> wrote:
>> On Tue, Apr 28, 2015 at 3:21 PM, Phil Pokorny
>> <[email protected]> wrote:
>>> On Tue, Apr 28, 2015 at 2:04 PM, Andy Lutomirski <[email protected]> wrote:
>>>> On Tue, Apr 28, 2015 at 11:25 AM, Dan Williams <[email protected]> wrote:
> [..]
>> This is such a mess that I think this driver should maybe flat-out
>> refuse to load in this type of configuration without some scary module
>> option. I have some NVDIMMs that report as type 12 but need two extra
>> out-of-tree drivers to work safely. First, they need i2c_imc or the
>> equivalent (I'll try to resubmit that soon). Second, they need secret
>> magic NDAed register poking. The latter is very problematic.
>>
>> At the very least, I think we should discourage people who don't
>> really know what they're doing from using this driver without care.
>
> The benefit of the type-12 experiment having not made it very far out
> of the lab is that it may be feasible to whitelist known platforms
> where we believe ADR is available. Otherwise, the presence of the
> NFIT asserts platform persistent memory support.
This could be a good idea. I'm planning on resubmitting my i2c driver
in the next couple weeks, and maybe I'll whitelist my own platform :)
--Andy
--
Andy Lutomirski
AMA Capital Management, LLC
On Wed, 2015-04-29 at 11:59 -0700, Dan Williams wrote:
> On Wed, Apr 29, 2015 at 11:25 AM, Toshi Kani <[email protected]> wrote:
> > Hi Dan,
> >
> > Thanks for the update. This version of the patchset enumerates our NFIT
> > table properly. :-)
> >
> > On Tue, 2015-04-28 at 14:25 -0400, Dan Williams wrote:
> >> In preparation for the pmem driver attaching to pmem-namespaces emitted
> >> by libnd, convert it to use an ida instead of an always increasing
> >> atomic index. This provides a bit of stability to pmem device names in
> >> the presence of driver re-bind events.
> > :
> >> @@ -122,20 +123,26 @@ static struct pmem_device *pmem_alloc(struct device *dev, struct resource *res)
> >> {
> >> struct pmem_device *pmem;
> >> struct gendisk *disk;
> >> - int idx, err;
> >> + int err;
> >>
> >> err = -ENOMEM;
> >> pmem = kzalloc(sizeof(*pmem), GFP_KERNEL);
> >> if (!pmem)
> >> goto out;
> >>
> >> + pmem->id = ida_simple_get(&pmem_ida, 0, 0, GFP_KERNEL);
> >
> > nd_pmem_probe() is called asynchronously via async_schedule_domain
> > (). We have seen a case that the region#->pmem# binding becomes
> > inconsistent across a reboot when there are 8 NVDIMM cards (reported by
> > Robert Elliott). This leads user to access a wrong device.
> >
> > I think pmem id needs to be assigned before async_schedule_domain(), and
> > cascaded to nd_pmem_probe().
> >
>
> I'll take a look at making this better, but it will never be
> bulletproof. For the same reason that root=UUID=<uuid> is preferred
> over root=/dev/sda userspace should never rely on consistent pmem
> device names from boot to boot.
I agree that constant unique IDs, such as UUIDs, are necessary to
guarantee their consistent numbering regardless of configuration
changes. For now, /dev/pmem%d should have consistent numbering while
NFIT table entries are consistent.
Thanks,
-Toshi
On Wed, Apr 29, 2015 at 11:25 AM, Toshi Kani <[email protected]> wrote:
> Hi Dan,
>
> Thanks for the update. This version of the patchset enumerates our NFIT
> table properly. :-)
>
> On Tue, 2015-04-28 at 14:25 -0400, Dan Williams wrote:
>> In preparation for the pmem driver attaching to pmem-namespaces emitted
>> by libnd, convert it to use an ida instead of an always increasing
>> atomic index. This provides a bit of stability to pmem device names in
>> the presence of driver re-bind events.
> :
>> @@ -122,20 +123,26 @@ static struct pmem_device *pmem_alloc(struct device *dev, struct resource *res)
>> {
>> struct pmem_device *pmem;
>> struct gendisk *disk;
>> - int idx, err;
>> + int err;
>>
>> err = -ENOMEM;
>> pmem = kzalloc(sizeof(*pmem), GFP_KERNEL);
>> if (!pmem)
>> goto out;
>>
>> + pmem->id = ida_simple_get(&pmem_ida, 0, 0, GFP_KERNEL);
>
> nd_pmem_probe() is called asynchronously via async_schedule_domain
> (). We have seen a case that the region#->pmem# binding becomes
> inconsistent across a reboot when there are 8 NVDIMM cards (reported by
> Robert Elliott). This leads user to access a wrong device.
>
> I think pmem id needs to be assigned before async_schedule_domain(), and
> cascaded to nd_pmem_probe().
>
I'll take a look at making this better, but it will never be
bulletproof. For the same reason that root=UUID=<uuid> is preferred
over root=/dev/sda userspace should never rely on consistent pmem
device names from boot to boot.
On Wed, Apr 29, 2015 at 10:10 AM, Dan Williams <[email protected]> wrote:
> On Tue, Apr 28, 2015 at 4:06 PM, Andy Lutomirski <[email protected]> wrote:
>> On Tue, Apr 28, 2015 at 3:30 PM, Dan Williams <[email protected]> wrote:
>>> On Tue, Apr 28, 2015 at 2:10 PM, Andy Lutomirski <[email protected]> wrote:
>>>> On Tue, Apr 28, 2015 at 11:26 AM, Dan Williams <[email protected]> wrote:
>>>>> From: Ross Zwisler <[email protected]>
>>>>>
>>>>> The libnd implementation handles allocating dimm address space (DPA)
>>>>> between PMEM and BLK mode interfaces. After DPA has been allocated from
>>>>> a BLK-region to a BLK-namespace the nd_blk driver attaches to handle I/O
>>>>> as a struct bio based block device. Unlike PMEM, BLK is required to
>>>>> handle platform specific details like mmio register formats and memory
>>>>> controller interleave. For this reason the libnd generic nd_blk driver
>>>>> calls back into the bus provider to carry out the I/O.
>>>>>
>>>>> This initial implementation handles the BLK interface defined by the
>>>>> ACPI 6 NFIT [1] and the NVDIMM DSM Interface Example [2] composed from
>>>>> DCR (dimm control region), BDW (block data window), IDT (interleave
>>>>> descriptor) NFIT structures and the hardware register format.
>>>>> [1]: http://www.uefi.org/sites/default/files/resources/ACPI_6.0.pdf
>>>>> [2]: http://pmem.io/documents/NVDIMM_DSM_Interface_Example.pdf
>>>>>
>>>>> Cc: Andy Lutomirski <[email protected]>
>>>>> Cc: Boaz Harrosh <[email protected]>
>>>>> Cc: H. Peter Anvin <[email protected]>
>>>>> Cc: Jens Axboe <[email protected]>
>>>>> Cc: Ingo Molnar <[email protected]>
>>>>> Cc: Christoph Hellwig <[email protected]>
>>>>> Signed-off-by: Ross Zwisler <[email protected]>
>>>>> Signed-off-by: Dan Williams <[email protected]>
>>>>> ---
>>>>> drivers/block/nd/Kconfig | 12 +
>>>>> drivers/block/nd/Makefile | 3
>>>>> drivers/block/nd/acpi.c | 422 +++++++++++++++++++++++++++++++++++--
>>>>> drivers/block/nd/acpi_nfit.h | 47 ++++
>>>>> drivers/block/nd/blk.c | 264 +++++++++++++++++++++++
>>>>> drivers/block/nd/libnd.h | 11 +
>>>>> drivers/block/nd/namespace_devs.c | 47 ++++
>>>>> drivers/block/nd/nd-private.h | 3
>>>>> drivers/block/nd/nd.h | 16 +
>>>>> drivers/block/nd/region.c | 8 +
>>>>> drivers/block/nd/region_devs.c | 65 +++++-
>>>>> drivers/block/nd/test/nfit.c | 29 +++
>>>>> drivers/block/nd/test/nfit_test.h | 2
>>>>> 13 files changed, 891 insertions(+), 38 deletions(-)
>>>>> create mode 100644 drivers/block/nd/blk.c
>>>>>
>>>>> diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
>>>>> index 612bf2b14283..bac4290129fc 100644
>>>>> --- a/drivers/block/nd/Kconfig
>>>>> +++ b/drivers/block/nd/Kconfig
>>>>> @@ -95,6 +95,18 @@ config BLK_DEV_PMEM
>>>>>
>>>>> Say Y if you want to use a NVDIMM described by ACPI, E820, etc...
>>>>>
>>>>> +config ND_BLK
>>>>> + tristate "BLK: Block data window (aperture) device support"
>>>>> + depends on LIBND
>>>>> + default ND_ACPI
>>>>> + help
>>>>> + This driver performs I/O using a set of mmio windows on a
>>>>> + dimm. The set of apertures will all access the one DIMM.
>>>>> + Multiple windows allow multiple threads to have a different
>>>>> + portions of the dimm open at one time.
>>>>> +
>>>>> + Say Y if you want to use a NVDIMM with BLK-mode capability
>>>>> +
>>>>
>>>> This describes how it works, not what it is. How about:
>>>>
>>>> This driver exposes NVDIMM BLK regions as block devices. BLK regions
>>>> are regions of NVDIMM storage that are sector-addressable, not
>>>> byte-addressible, and do not support DAX.
>>>
>>> They *are* byte-addressable albeit through an indirection window. The
>>> indirection windows are too small for DAX to be a viable access mode.
>>
>> Right, I was assuming incorrectly that the sector-atomic thing was a
>> necessary part of BLK, or at least of this implementation.
>>
>> Anyway, I think my point stands: let's describe what these drivers do
>> from a user's perspective, not how they work.
>
> Agreed. How about:
>
> "Support NVDIMMs, or other devices, that implement a BLK-mode access
> capability. BLK-mode access uses memory-mapped-i/o apertures to
> access persistent media.
>
> Say Y if your platform firmware emits an ACPI.NFIT table
> (CONFIG_ND_ACPI), or otherwise exposes BLK-mode capabilities."
Sure.
--Andy
On 4/29/2015 2:53 PM, Toshi Kani wrote:
> On Wed, 2015-04-29 at 11:59 -0700, Dan Williams wrote:
>> On Wed, Apr 29, 2015 at 11:25 AM, Toshi Kani <[email protected]> wrote:
>>> Hi Dan,
>>>
>>> Thanks for the update. This version of the patchset enumerates our NFIT
>>> table properly. :-)
>>>
>>> On Tue, 2015-04-28 at 14:25 -0400, Dan Williams wrote:
>>>> In preparation for the pmem driver attaching to pmem-namespaces emitted
>>>> by libnd, convert it to use an ida instead of an always increasing
>>>> atomic index. This provides a bit of stability to pmem device names in
>>>> the presence of driver re-bind events.
>>> :
>>>> @@ -122,20 +123,26 @@ static struct pmem_device *pmem_alloc(struct device *dev, struct resource *res)
>>>> {
>>>> struct pmem_device *pmem;
>>>> struct gendisk *disk;
>>>> - int idx, err;
>>>> + int err;
>>>>
>>>> err = -ENOMEM;
>>>> pmem = kzalloc(sizeof(*pmem), GFP_KERNEL);
>>>> if (!pmem)
>>>> goto out;
>>>>
>>>> + pmem->id = ida_simple_get(&pmem_ida, 0, 0, GFP_KERNEL);
>>>
>>> nd_pmem_probe() is called asynchronously via async_schedule_domain
>>> (). We have seen a case that the region#->pmem# binding becomes
>>> inconsistent across a reboot when there are 8 NVDIMM cards (reported by
>>> Robert Elliott). This leads user to access a wrong device.
>>>
>>> I think pmem id needs to be assigned before async_schedule_domain(), and
>>> cascaded to nd_pmem_probe().
>>>
>>
>> I'll take a look at making this better, but it will never be
>> bulletproof. For the same reason that root=UUID=<uuid> is preferred
>> over root=/dev/sda userspace should never rely on consistent pmem
>> device names from boot to boot.
>
> I agree that constant unique IDs, such as UUIDs, are necessary to
> guarantee their consistent numbering regardless of configuration
> changes. For now, /dev/pmem%d should have consistent numbering while
> NFIT table entries are consistent.
What's the right answer for this in the long run?
The NFIT has information like "Memory Device Physical ID" that is an SMBIOS
type 17 handle. SMBIOS could have some naming information, like it does for
NICs. There is also a "Memory Device Region ID". Is that combination
enough to give us a unique identifier?
Of course, I'm assuming that information is consistent across reboots
and ideally across configuration changes, like adding more NVDIMMs.
For NICs, that's where the SMBIOS consistent naming information is helpful.
Is that what we need for devices that don't have namespaces and labels?
-- ljk
>
> Thanks,
> -Toshi
>
> _______________________________________________
> Linux-nvdimm mailing list
> [email protected]
> https://lists.01.org/mailman/listinfo/linux-nvdimm
>
On Wed, Apr 29, 2015 at 1:49 PM, Linda Knippers <[email protected]> wrote:
> On 4/29/2015 2:53 PM, Toshi Kani wrote:
> What's the right answer for this in the long run?
Short term, /dev/disk/by-uuid to take a stable identifier from the
contents of the device.
Longer term teach udev to populate /dev/disk/by-id with stable names
for libnd devices. The trick is identifiers for interleaved PMEM
ranges comprised of multiple physical devices. I'm thinking something
like /dev/disk/by-id/nd-<set_cookie>
On Tue, 2015-04-28 at 16:05 -0700, Andy Lutomirski wrote:
> On Tue, Apr 28, 2015 at 3:28 PM, Dan Williams <[email protected]> wrote:
> > On Tue, Apr 28, 2015 at 2:06 PM, Andy Lutomirski <[email protected]> wrote:
> >> On Tue, Apr 28, 2015 at 1:59 PM, Dan Williams <[email protected]> wrote:
> >>> On Tue, Apr 28, 2015 at 1:52 PM, Andy Lutomirski <[email protected]> wrote:
> >>>> On Tue, Apr 28, 2015 at 11:24 AM, Dan Williams <[email protected]> wrote:
> >>>> Mostly for my understanding: is there a name for "address relative to
> >>>> the address lines on the DIMM"? That is, a DIMM that exposes 8 GB of
> >>>> apparent physical memory, possibly interleaved, broken up, or weirdly
> >>>> remapped by the memory controller, would still have addresses between
> >>>> 0 and 8 GB. Some of those might be PMEM windows, some might be MMIO,
> >>>> some might be BLK apertures, etc.
> >>>>
> >>>> IIUC "DPA" refers to actual addressable storage, not this type of address?
> >>>
> >>> No, DPA is exactly as you describe above. You can't directly access
> >>> it except through a PMEM mapping (possibly interleaved with DPA from
> >>> other DIMMs) or a BLK aperture (mmio window into DPA).
> >>
> >> So the thing I'm describing has no name, then? Oh, well.
> >
> > What? The thing you are describing *is* DPA.
>
> I'm confused. Here are the two things I have in mind:
>
> 1. An address into on-DIMM storage. If I have a DIMM that is mapped
> to 8 GB of SPA but has 64 GB of usable storage (accessed through BLK
> apertures, say), then this address runs from 0 to 64 GB.
>
> 2. An address into the DIMM's view of physical address space. If I
> have a DIMM that is mapped to 8 GB of SPA but has 64 GB of usable
> storage (accessed through BLK apertures, say), then this address runs
> from 0 to 8 GB. There's a one-to-one mapping between SPA and this
> type of address.
>
> Since you said "a dimm may provide both PMEM-mode and BLK-mode access
> to a range of DPA.," I thought that DPA was #1.
>
> --Andy
I think that you've got the right definition, #1 above, for DPA. The DPA is
relative to the DIMM, knows nothing about interleaving or SPA or anything else
in the system, and is basically equivalent to the idea of an LBA on a disk. A
DIMM that has 64 GiB of storage could have a DPA space ranging from 0 to 64
GiB.
The second concept is a little trickier - we've been talking about this by
using the term "N-way interleave set". Say you have your 64 GiB DIMM and only
the first 8 GiB are given to the OS in an SPA, and that DIMM isn't interleaved
with any other DIMMs. This would be a 1-way interleave set, ranging from DPA
0 - 8GiB on the DIMM.
If you have 2 DIMMs of size 64 GiB, and they each have a 8 GiB region given to
the SPA space, those two regions could be interleaved together. The OS would
then see a 16 GiB 2-way interleave set, made up of DPAs 0 -> 8 GiB on each of
the two DIMMs.
You can figure out exactly how all the interleaving works by looking at the
SPA tables, the Memory Device tables and the Interleave Tables.
These are in sections 5.2.25.1 - 5.2.25.3 in ACPI 6, and are in our code as
struct acpi_nfit_spa, struct acpi_nfit_memdev and struct acpi_nfit_idt.
- Ross
On Tuesday, April 28, 2015 02:24:23 PM Dan Williams wrote:
> 1/ Autodetect an NFIT table for the ACPI namespace device with _HID of
> "ACPI0012"
>
> 2/ libnd bus registration
>
> The NFIT provided by ACPI is one possible method by which platforms will
> discover NVDIMM resources. However, the intent of the nd_bus_descriptor
> abstraction is to abstract "provider" specific details, leaving libnd
> to be independent of the specific NVDIMM resource discovery mechanism.
> This flexibility is later exploited later to implement custom-defined nd
> buses.
>
> Cc: <[email protected]>
> Cc: Robert Moore <[email protected]>
> Cc: Rafael J. Wysocki <[email protected]>
> Signed-off-by: Dan Williams <[email protected]>
> ---
> drivers/block/Kconfig | 2
> drivers/block/Makefile | 1
> drivers/block/nd/Kconfig | 40 +++
> drivers/block/nd/Makefile | 6 +
> drivers/block/nd/acpi.c | 475 +++++++++++++++++++++++++++++++++++++++++
> drivers/block/nd/acpi_nfit.h | 254 ++++++++++++++++++++++
> drivers/block/nd/core.c | 67 ++++++
> drivers/block/nd/libnd.h | 33 +++
> drivers/block/nd/nd-private.h | 23 ++
> 9 files changed, 901 insertions(+)
> create mode 100644 drivers/block/nd/Kconfig
> create mode 100644 drivers/block/nd/Makefile
> create mode 100644 drivers/block/nd/acpi.c
> create mode 100644 drivers/block/nd/acpi_nfit.h
> create mode 100644 drivers/block/nd/core.c
> create mode 100644 drivers/block/nd/libnd.h
> create mode 100644 drivers/block/nd/nd-private.h
>
> diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig
> index eb1fed5bd516..dfe40e5ca9bd 100644
> --- a/drivers/block/Kconfig
> +++ b/drivers/block/Kconfig
> @@ -321,6 +321,8 @@ config BLK_DEV_NVME
> To compile this driver as a module, choose M here: the
> module will be called nvme.
>
> +source "drivers/block/nd/Kconfig"
> +
> config BLK_DEV_SKD
> tristate "STEC S1120 Block Driver"
> depends on PCI
> diff --git a/drivers/block/Makefile b/drivers/block/Makefile
> index 9cc6c18a1c7e..07a6acecf4d8 100644
> --- a/drivers/block/Makefile
> +++ b/drivers/block/Makefile
> @@ -24,6 +24,7 @@ obj-$(CONFIG_CDROM_PKTCDVD) += pktcdvd.o
> obj-$(CONFIG_MG_DISK) += mg_disk.o
> obj-$(CONFIG_SUNVDC) += sunvdc.o
> obj-$(CONFIG_BLK_DEV_NVME) += nvme.o
> +obj-$(CONFIG_ND_DEVICES) += nd/
> obj-$(CONFIG_BLK_DEV_SKD) += skd.o
> obj-$(CONFIG_BLK_DEV_OSD) += osdblk.o
>
> diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
> new file mode 100644
> index 000000000000..6d5d6b732f82
> --- /dev/null
> +++ b/drivers/block/nd/Kconfig
> @@ -0,0 +1,40 @@
> +menuconfig ND_DEVICES
> + bool "NVDIMM Support"
> + depends on PHYS_ADDR_T_64BIT
> + help
> + Generic support for non-volatile memory devices including
> + ACPI-6-NFIT defined resources. On platforms that define an
> + NFIT, or otherwise can discover NVDIMM resources, a libnd
> + bus is registered to advertise PMEM (persistent memory)
> + namespaces (/dev/pmemX) and BLK (sliding mmio window(s))
> + namespaces (/dev/ndX). A PMEM namespace refers to a memory
> + resource that may span multiple DIMMs and support DAX (see
> + CONFIG_DAX). A BLK namespace refers to an NVDIMM control
> + region which exposes an mmio register set for windowed
> + access mode to non-volatile memory.
> +
> +if ND_DEVICES
> +
> +config LIBND
> + tristate "LIBND: libnd device driver support"
> + help
> + Platform agnostic device model for a libnd bus. Publishes
> + resources for a PMEM (persistent-memory) driver and/or BLK
> + (sliding mmio window(s)) driver to attach. Exposes a device
> + topology under a "ndX" bus device, a "/dev/ndctlX" bus-ioctl
> + message passing interface, and a "/dev/nmemX" dimm-ioctl
> + message interface for each memory device registered on the
> + bus. instance. A userspace library "ndctl" provides an API
> + to enumerate/manage this subsystem.
> +
> +config ND_ACPI
> + tristate "ACPI: NFIT to libnd bus support"
> + select LIBND
> + depends on ACPI
> + help
> + Infrastructure to probe ACPI 6 compliant platforms for
> + NVDIMMs (NFIT) and register a libnd device tree. In
> + addition to storage devices this also enables libnd craft
> + ACPI._DSM messages for platform/dimm configuration.
I'm wondering if the two CONFIG options above really need to be user-selectable?
For example, what reason people (who've already selected ND_DEVICES) may have
for not selecting ND_ACPI if ACPI is set?
> +
> +endif
> diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
> new file mode 100644
> index 000000000000..944b5947c0cb
> --- /dev/null
> +++ b/drivers/block/nd/Makefile
> @@ -0,0 +1,6 @@
> +obj-$(CONFIG_LIBND) += libnd.o
> +obj-$(CONFIG_ND_ACPI) += nd_acpi.o
> +
> +nd_acpi-y := acpi.o
> +
> +libnd-y := core.o
OK, so it looks like no modules, just built-in code, right?
> diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
> new file mode 100644
> index 000000000000..9f0b24390d1b
> --- /dev/null
> +++ b/drivers/block/nd/acpi.c
> @@ -0,0 +1,475 @@
> +/*
> + * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of version 2 of the GNU General Public License as
> + * published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope that it will be useful, but
> + * WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> + * General Public License for more details.
> + */
> +#include <linux/list_sort.h>
> +#include <linux/module.h>
> +#include <linux/list.h>
> +#include <linux/acpi.h>
> +#include "acpi_nfit.h"
> +#include "libnd.h"
> +
> +static bool warn_checksum;
> +module_param(warn_checksum, bool, S_IRUGO|S_IWUSR);
> +MODULE_PARM_DESC(warn_checksum, "Turn checksum errors into warnings");
> +
> +enum {
> + NFIT_ACPI_NOTIFY_TABLE = 0x80,
> +};
> +
> +static int nd_acpi_ctl(struct nd_bus_descriptor *nd_desc,
> + struct nd_dimm *nd_dimm, unsigned int cmd, void *buf,
> + unsigned int buf_len)
> +{
> + return -ENOTTY;
> +}
Why -ENOTTY? And why not to leave a NULL entry for this instead and
make the library fail it in that case?
> +
> +static const char *spa_type_name(u16 type)
> +{
> + switch (type) {
> + case NFIT_SPA_VOLATILE: return "volatile";
> + case NFIT_SPA_PM: return "pmem";
> + case NFIT_SPA_DCR: return "dimm-control-region";
> + case NFIT_SPA_BDW: return "block-data-window";
> + default: return "unknown";
> + }
> +}
> +
> +static int nfit_spa_type(struct acpi_nfit_spa *spa)
> +{
> + if (memcmp(&nfit_spa_uuid_volatile, spa->type_uuid, 16) == 0)
> + return NFIT_SPA_VOLATILE;
> +
> + if (memcmp(&nfit_spa_uuid_pm, spa->type_uuid, 16) == 0)
> + return NFIT_SPA_PM;
> +
> + if (memcmp(&nfit_spa_uuid_dcr, spa->type_uuid, 16) == 0)
> + return NFIT_SPA_DCR;
> +
> + if (memcmp(&nfit_spa_uuid_bdw, spa->type_uuid, 16) == 0)
> + return NFIT_SPA_BDW;
> +
> + if (memcmp(&nfit_spa_uuid_vdisk, spa->type_uuid, 16) == 0)
> + return NFIT_SPA_VDISK;
> +
> + if (memcmp(&nfit_spa_uuid_vcd, spa->type_uuid, 16) == 0)
> + return NFIT_SPA_VCD;
> +
> + if (memcmp(&nfit_spa_uuid_pdisk, spa->type_uuid, 16) == 0)
> + return NFIT_SPA_PDISK;
> +
> + if (memcmp(&nfit_spa_uuid_pcd, spa->type_uuid, 16) == 0)
> + return NFIT_SPA_PCD;
> +
> + return -1;
> +}
> +
> +struct nfit_table_header {
> + __le16 type;
> + __le16 length;
> +};
That you'll be able to get from ACPICA I suppose?
> +
> +static void *add_table(struct acpi_nfit_desc *acpi_desc, void *table, const void *end)
> +{
> + struct device *dev = acpi_desc->dev;
> + struct nfit_table_header *hdr;
> + void *err = ERR_PTR(-ENOMEM);
> +
> + if (table >= end)
> + return NULL;
> +
> + hdr = (struct nfit_table_header *) table;
> + switch (hdr->type) {
> + case NFIT_TABLE_SPA: {
> + struct nfit_spa *nfit_spa = devm_kzalloc(dev, sizeof(*nfit_spa),
> + GFP_KERNEL);
> + struct acpi_nfit_spa *spa = table;
> +
> + if (!nfit_spa)
> + return err;
> + INIT_LIST_HEAD(&nfit_spa->list);
> + nfit_spa->spa = spa;
> + list_add_tail(&nfit_spa->list, &acpi_desc->spas);
> + dev_dbg(dev, "%s: spa index: %d type: %s\n", __func__,
> + spa->spa_index,
> + spa_type_name(nfit_spa_type(spa)));
> + break;
> + }
> + case NFIT_TABLE_MEM: {
> + struct nfit_memdev *nfit_memdev = devm_kzalloc(dev,
> + sizeof(*nfit_memdev), GFP_KERNEL);
> + struct acpi_nfit_memdev *memdev = table;
> +
> + if (!nfit_memdev)
> + return err;
> + INIT_LIST_HEAD(&nfit_memdev->list);
> + nfit_memdev->memdev = memdev;
> + list_add_tail(&nfit_memdev->list, &acpi_desc->memdevs);
> + dev_dbg(dev, "%s: memdev handle: %#x spa: %d dcr: %d\n",
> + __func__, memdev->nfit_handle, memdev->spa_index,
> + memdev->dcr_index);
> + break;
> + }
> + case NFIT_TABLE_DCR: {
> + struct nfit_dcr *nfit_dcr = devm_kzalloc(dev, sizeof(*nfit_dcr),
> + GFP_KERNEL);
> + struct acpi_nfit_dcr *dcr = table;
> +
> + if (!nfit_dcr)
> + return err;
> + INIT_LIST_HEAD(&nfit_dcr->list);
> + nfit_dcr->dcr = dcr;
> + list_add_tail(&nfit_dcr->list, &acpi_desc->dcrs);
> + dev_dbg(dev, "%s: dcr index: %d num_bcw: %d\n", __func__,
> + dcr->dcr_index, dcr->num_bcw);
> + break;
> + }
> + case NFIT_TABLE_BDW: {
> + struct nfit_bdw *nfit_bdw = devm_kzalloc(dev, sizeof(*nfit_bdw),
> + GFP_KERNEL);
> + struct acpi_nfit_bdw *bdw = table;
> +
> + if (!nfit_bdw)
> + return err;
> + INIT_LIST_HEAD(&nfit_bdw->list);
> + nfit_bdw->bdw = bdw;
> + list_add_tail(&nfit_bdw->list, &acpi_desc->bdws);
> + dev_dbg(dev, "%s: bdw dcr: %d num_bdw: %d\n", __func__,
> + bdw->dcr_index, bdw->num_bdw);
> + break;
> + }
> + /* TODO */
> + case NFIT_TABLE_IDT:
> + dev_dbg(dev, "%s: idt\n", __func__);
> + break;
> + case NFIT_TABLE_FLUSH:
> + dev_dbg(dev, "%s: flush\n", __func__);
> + break;
> + case NFIT_TABLE_SMBIOS:
> + dev_dbg(dev, "%s: smbios\n", __func__);
> + break;
> + default:
> + dev_err(dev, "unknown table '%d' parsing nfit\n", hdr->type);
> + return ERR_PTR(-ENXIO);
> + }
> +
> + return table + hdr->length;
> +}
> +
> +static void nfit_mem_find_spa_bdw(struct acpi_nfit_desc *acpi_desc,
> + struct nfit_mem *nfit_mem)
> +{
> + u32 nfit_handle = __to_nfit_memdev(nfit_mem)->nfit_handle;
> + u16 dcr_index = nfit_mem->dcr->dcr_index;
> + struct nfit_spa *nfit_spa;
> +
> + list_for_each_entry(nfit_spa, &acpi_desc->spas, list) {
> + u16 spa_index = nfit_spa->spa->spa_index;
> + int type = nfit_spa_type(nfit_spa->spa);
> + struct nfit_memdev *nfit_memdev;
> +
> + if (type != NFIT_SPA_BDW)
> + continue;
> +
> + list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) {
> + if (nfit_memdev->memdev->spa_index != spa_index)
> + continue;
> + if (nfit_memdev->memdev->nfit_handle != nfit_handle)
> + continue;
> + if (nfit_memdev->memdev->dcr_index != dcr_index)
> + continue;
> +
> + nfit_mem->spa_bdw = nfit_spa->spa;
> + return;
> + }
> + }
> +
> + dev_dbg(acpi_desc->dev, "SPA-BDW not found for SPA-DCR %d\n",
> + nfit_mem->spa_dcr->spa_index);
> + nfit_mem->bdw = NULL;
> +}
> +
> +static int nfit_mem_add(struct acpi_nfit_desc *acpi_desc,
> + struct nfit_mem *nfit_mem, struct acpi_nfit_spa *spa)
> +{
> + u16 dcr_index = __to_nfit_memdev(nfit_mem)->dcr_index;
> + struct nfit_dcr *nfit_dcr;
> + struct nfit_bdw *nfit_bdw;
> +
> + list_for_each_entry(nfit_dcr, &acpi_desc->dcrs, list) {
> + if (nfit_dcr->dcr->dcr_index != dcr_index)
> + continue;
> + nfit_mem->dcr = nfit_dcr->dcr;
> + break;
> + }
> +
> + if (!nfit_mem->dcr) {
> + dev_dbg(acpi_desc->dev, "SPA %d missing:%s%s\n", spa->spa_index,
> + __to_nfit_memdev(nfit_mem) ? "" : " MEMDEV",
> + nfit_mem->dcr ? "" : " DCR");
> + return -ENODEV;
> + }
> +
> + /*
> + * We've found enough to create an nd_dimm, optionally
> + * find an associated BDW
> + */
> + list_add(&nfit_mem->list, &acpi_desc->dimms);
> +
> + list_for_each_entry(nfit_bdw, &acpi_desc->bdws, list) {
> + if (nfit_bdw->bdw->dcr_index != dcr_index)
> + continue;
> + nfit_mem->bdw = nfit_bdw->bdw;
> + break;
> + }
> +
> + if (!nfit_mem->bdw)
> + return 0;
> +
> + nfit_mem_find_spa_bdw(acpi_desc, nfit_mem);
> + return 0;
> +}
> +
> +static int nfit_mem_dcr_init(struct acpi_nfit_desc *acpi_desc,
> + struct acpi_nfit_spa *spa)
> +{
> + struct nfit_mem *nfit_mem, *found;
> + struct nfit_memdev *nfit_memdev;
> + int type = nfit_spa_type(spa);
> + u16 dcr_index;
> +
> + switch (type) {
> + case NFIT_SPA_DCR:
> + case NFIT_SPA_PM:
> + break;
> + default:
> + return 0;
> + }
> +
> + list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) {
> + int rc;
> +
> + if (nfit_memdev->memdev->spa_index != spa->spa_index)
> + continue;
> + found = NULL;
> + dcr_index = nfit_memdev->memdev->dcr_index;
> + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list)
> + if (__to_nfit_memdev(nfit_mem)->dcr_index == dcr_index) {
> + found = nfit_mem;
> + break;
> + }
> +
> + if (found)
> + nfit_mem = found;
> + else {
> + nfit_mem = devm_kzalloc(acpi_desc->dev,
> + sizeof(*nfit_mem), GFP_KERNEL);
> + if (!nfit_mem)
> + return -ENOMEM;
> + INIT_LIST_HEAD(&nfit_mem->list);
> + }
> +
> + if (type == NFIT_SPA_DCR) {
> + /* multiple dimms may share a SPA when interleaved */
> + nfit_mem->spa_dcr = spa;
> + nfit_mem->memdev_dcr = nfit_memdev->memdev;
> + } else {
> + /*
> + * A single dimm may belong to multiple SPA-PM
> + * ranges, record at least one in addition to
> + * any SPA-DCR range.
> + */
> + nfit_mem->memdev_pmem = nfit_memdev->memdev;
> + }
> +
> + if (found)
> + continue;
> +
> + rc = nfit_mem_add(acpi_desc, nfit_mem, spa);
> + if (rc)
> + return rc;
> + }
> +
> + return 0;
> +}
> +
> +static int nfit_mem_cmp(void *priv, struct list_head *__a, struct list_head *__b)
> +{
> + struct nfit_mem *a = container_of(__a, typeof(*a), list);
> + struct nfit_mem *b = container_of(__b, typeof(*b), list);
> + u32 handleA, handleB;
> +
> + handleA = __to_nfit_memdev(a)->nfit_handle;
> + handleB = __to_nfit_memdev(b)->nfit_handle;
> + if (handleA < handleB)
> + return -1;
> + else if (handleA > handleB)
> + return 1;
> + return 0;
> +}
> +
> +static int nfit_mem_init(struct acpi_nfit_desc *acpi_desc)
> +{
> + struct nfit_spa *nfit_spa;
> +
> + /*
> + * For each SPA-DCR or SPA-PMEM address range find its
> + * corresponding MEMDEV(s). From each MEMDEV find the
> + * corresponding DCR. Then, if we're operating on a SPA-DCR,
> + * try to find a SPA-BDW and a corresponding BDW that references
> + * the DCR. Throw it all into an nfit_mem object. Note, that
> + * BDWs are optional.
> + */
> + list_for_each_entry(nfit_spa, &acpi_desc->spas, list) {
> + int rc;
> +
> + rc = nfit_mem_dcr_init(acpi_desc, nfit_spa->spa);
> + if (rc)
> + return rc;
> + }
> +
> + list_sort(NULL, &acpi_desc->dimms, nfit_mem_cmp);
> +
> + return 0;
> +}
> +
> +static int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
> +{
> + struct device *dev = acpi_desc->dev;
> + const void *end;
> + u8 *data, sum;
> + acpi_size i;
> +
> + INIT_LIST_HEAD(&acpi_desc->spas);
> + INIT_LIST_HEAD(&acpi_desc->dcrs);
> + INIT_LIST_HEAD(&acpi_desc->bdws);
> + INIT_LIST_HEAD(&acpi_desc->memdevs);
> + INIT_LIST_HEAD(&acpi_desc->dimms);
> +
> + data = (u8 *) acpi_desc->nfit;
> + for (i = 0, sum = 0; i < sz; i++)
> + sum += readb(data + i);
> + if (sum != 0 && !warn_checksum) {
> + dev_dbg(dev, "%s: nfit checksum failure\n", __func__);
> + return -ENXIO;
> + }
> + WARN_TAINT_ONCE(sum != 0, TAINT_FIRMWARE_WORKAROUND,
> + "nfit checksum failure, continuing...\n");
> +
> + end = data + sz;
> + data += sizeof(struct acpi_nfit);
> + while (!IS_ERR_OR_NULL(data))
> + data = add_table(acpi_desc, data, end);
This looks like we are expecting a series of tables here and we're going to
fail the whole discovery if just one of them in invalid.
I wonder if it would be practical to skip just the invalid ones instead?
> +
> + if (IS_ERR(data)) {
> + dev_dbg(dev, "%s: nfit table parsing error: %ld\n", __func__,
> + PTR_ERR(data));
> + return PTR_ERR(data);
> + }
> +
> + if (nfit_mem_init(acpi_desc) != 0)
> + return -ENOMEM;
> +
> + return 0;
> +}
> +
> +static int nd_acpi_add(struct acpi_device *adev)
> +{
> + struct nd_bus_descriptor *nd_desc;
> + struct acpi_nfit_desc *acpi_desc;
> + struct device *dev = &adev->dev;
> + struct acpi_table_header *tbl;
> + acpi_status status = AE_OK;
> + acpi_size sz;
> + int rc;
> +
> + status = acpi_get_table_with_size("NFIT", 0, &tbl, &sz);
> + if (ACPI_FAILURE(status)) {
> + dev_err(dev, "failed to find NFIT\n");
> + return -ENXIO;
> + }
> +
> + acpi_desc = devm_kzalloc(dev, sizeof(*acpi_desc), GFP_KERNEL);
> + if (!acpi_desc)
> + return -ENOMEM;
> +
> + dev_set_drvdata(dev, acpi_desc);
> + acpi_desc->dev = dev;
> + acpi_desc->nfit = (struct acpi_nfit *) tbl;
> + nd_desc = &acpi_desc->nd_desc;
> + nd_desc->provider_name = "ACPI.NFIT";
> + nd_desc->ndctl = nd_acpi_ctl;
> +
> + acpi_desc->nd_bus = nd_bus_register(dev, nd_desc);
> + if (!acpi_desc->nd_bus)
> + return -ENXIO;
> +
> + rc = nd_acpi_nfit_init(acpi_desc, sz);
> + if (rc) {
> + nd_bus_unregister(acpi_desc->nd_bus);
> + return rc;
> + }
> + return 0;
> +}
> +
> +static int nd_acpi_remove(struct acpi_device *adev)
> +{
> + struct acpi_nfit_desc *acpi_desc = dev_get_drvdata(&adev->dev);
> +
> + nd_bus_unregister(acpi_desc->nd_bus);
> + return 0;
> +}
> +
> +static void nd_acpi_notify(struct acpi_device *adev, u32 event)
> +{
> + /* TODO: handle ACPI_NOTIFY_BUS_CHECK notification */
> + dev_dbg(&adev->dev, "%s: event: %d\n", __func__, event);
> +}
> +
> +static const struct acpi_device_id nd_acpi_ids[] = {
> + { "ACPI0012", 0 },
> + { "", 0 },
> +};
> +MODULE_DEVICE_TABLE(acpi, nd_acpi_ids);
> +
> +static struct acpi_driver nd_acpi_driver = {
> + .name = KBUILD_MODNAME,
> + .ids = nd_acpi_ids,
> + .flags = ACPI_DRIVER_ALL_NOTIFY_EVENTS,
> + .ops = {
> + .add = nd_acpi_add,
> + .remove = nd_acpi_remove,
> + .notify = nd_acpi_notify
> + },
> +};
Since this is going to be non-modular built-in code, please use an ACPI
scan handler instead of using a driver here. acpi_memhotplug.c does that,
you can use it as an example, but I guess you don't need to enable hotplug
for it to start with.
> +
> +static __init int nd_acpi_init(void)
> +{
> + BUILD_BUG_ON(sizeof(struct acpi_nfit) != 40);
> + BUILD_BUG_ON(sizeof(struct acpi_nfit_spa) != 56);
> + BUILD_BUG_ON(sizeof(struct acpi_nfit_memdev) != 48);
> + BUILD_BUG_ON(sizeof(struct acpi_nfit_idt) != 16);
> + BUILD_BUG_ON(sizeof(struct acpi_nfit_smbios) != 8);
> + BUILD_BUG_ON(sizeof(struct acpi_nfit_dcr) != 80);
> + BUILD_BUG_ON(sizeof(struct acpi_nfit_bdw) != 40);
> +
> + return acpi_bus_register_driver(&nd_acpi_driver);
> +}
> +
> +static __exit void nd_acpi_exit(void)
> +{
> + acpi_bus_unregister_driver(&nd_acpi_driver);
> +}
> +
> +module_init(nd_acpi_init);
> +module_exit(nd_acpi_exit);
> +MODULE_LICENSE("GPL v2");
> +MODULE_AUTHOR("Intel Corporation");
> diff --git a/drivers/block/nd/acpi_nfit.h b/drivers/block/nd/acpi_nfit.h
> new file mode 100644
> index 000000000000..e0b0f12736bf
> --- /dev/null
> +++ b/drivers/block/nd/acpi_nfit.h
I'm assuming that the below is coordinated with Bob and David and will be
changed to use ACPICA-provided definitions going forward.
Is that correct?
> @@ -0,0 +1,254 @@
> +/*
> + * NVDIMM Firmware Interface Table - NFIT
> + *
> + * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of version 2 of the GNU General Public License as
> + * published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope that it will be useful, but
> + * WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> + * General Public License for more details.
> + */
> +#ifndef __NFIT_H__
> +#define __NFIT_H__
> +#include <linux/types.h>
> +#include <linux/uuid.h>
> +#include <linux/acpi.h>
> +#include "libnd.h"
> +
> +static const uuid_le nfit_spa_uuid_volatile __maybe_unused = UUID_LE(0x7305944f,
> + 0xfdda, 0x44e3, 0xb1, 0x6c, 0x3f, 0x22, 0xd2, 0x52, 0xe5, 0xd0);
> +
> +static const uuid_le nfit_spa_uuid_pm __maybe_unused = UUID_LE(0x66f0d379,
> + 0xb4f3, 0x4074, 0xac, 0x43, 0x0d, 0x33, 0x18, 0xb7, 0x8c, 0xdb);
> +
> +static const uuid_le nfit_spa_uuid_dcr __maybe_unused = UUID_LE(0x92f701f6,
> + 0x13b4, 0x405d, 0x91, 0x0b, 0x29, 0x93, 0x67, 0xe8, 0x23, 0x4c);
> +
> +static const uuid_le nfit_spa_uuid_bdw __maybe_unused = UUID_LE(0x91af0530,
> + 0x5d86, 0x470e, 0xa6, 0xb0, 0x0a, 0x2d, 0xb9, 0x40, 0x82, 0x49);
> +
> +static const uuid_le nfit_spa_uuid_vdisk __maybe_unused = UUID_LE(0x77ab535a,
> + 0x45fc, 0x624b, 0x55, 0x60, 0xf7, 0xb2, 0x81, 0xd1, 0xf9, 0x6e);
> +
> +static const uuid_le nfit_spa_uuid_vcd __maybe_unused = UUID_LE(0x3d5abd30,
> + 0x4175, 0x87ce, 0x6d, 0x64, 0xd2, 0xad, 0xe5, 0x23, 0xc4, 0xbb);
> +
> +static const uuid_le nfit_spa_uuid_pdisk __maybe_unused = UUID_LE(0x5cea02c9,
> + 0x4d07, 0x69d3, 0x26, 0x9f, 0x44, 0x96, 0xfb, 0xe0, 0x96, 0xf9);
> +
> +static const uuid_le nfit_spa_uuid_pcd __maybe_unused = UUID_LE(0x08018188,
> + 0x42cd, 0xbb48, 0x10, 0x0f, 0x53, 0x87, 0xd5, 0x3d, 0xed, 0x3d);
> +
> +enum {
> + NFIT_TABLE_SPA = 0,
> + NFIT_TABLE_MEM = 1,
> + NFIT_TABLE_IDT = 2,
> + NFIT_TABLE_SMBIOS = 3,
> + NFIT_TABLE_DCR = 4,
> + NFIT_TABLE_BDW = 5,
> + NFIT_TABLE_FLUSH = 6,
> + NFIT_SPA_VOLATILE = 0,
> + NFIT_SPA_PM = 1,
> + NFIT_SPA_DCR = 2,
> + NFIT_SPA_BDW = 3,
> + NFIT_SPA_VDISK = 4,
> + NFIT_SPA_VCD = 5,
> + NFIT_SPA_PDISK = 6,
> + NFIT_SPA_PCD = 7,
> + NFIT_SPAF_DCR_HOT_ADD = 1 << 0,
> + NFIT_SPAF_PDVALID = 1 << 1,
> + NFIT_MEMF_SAVE_FAIL = 1 << 0,
> + NFIT_MEMF_RESTORE_FAIL = 1 << 1,
> + NFIT_MEMF_FLUSH_FAIL = 1 << 2,
> + NFIT_MEMF_UNARMED = 1 << 3,
> + NFIT_MEMF_NOTIFY_SMART = 1 << 4,
> + NFIT_MEMF_SMART_READY = 1 << 5,
> + NFIT_DCRF_BUFFERED = 1 << 0,
> +};
> +
> +/**
> + * struct acpi_nfit - Nvdimm Firmware Interface Table
> + * @signature: "NFIT"
> + * @length: sum of size of this table plus all appended subtables
> + */
> +struct acpi_nfit {
> + u8 signature[4];
> + u32 length;
> + u8 revision;
> + u8 checksum;
> + u8 oemid[6];
> + u64 oem_tbl_id;
> + u32 oem_revision;
> + u32 creator_id;
> + u32 creator_revision;
> + u32 reserved;
> +};
> +
> +/**
> + * struct acpi_nfit_spa - System Physical Address Range Descriptor Table
> + */
> +struct acpi_nfit_spa {
> + u16 type;
> + u16 length;
> + u16 spa_index;
> + u16 flags;
> + u32 reserved;
> + u32 proximity_domain;
> + u8 type_uuid[16];
> + u64 spa_base;
> + u64 spa_length;
> + u64 mem_attr;
> +};
> +
> +/**
> + * struct acpi_nfit_mem - Memory Device to SPA Mapping Table
> + */
> +struct acpi_nfit_memdev {
> + u16 type;
> + u16 length;
> + u32 nfit_handle;
> + u16 phys_id;
> + u16 region_id;
> + u16 spa_index;
> + u16 dcr_index;
> + u64 region_len;
> + u64 region_spa_offset;
> + u64 region_dpa;
> + u16 idt_index;
> + u16 interleave_ways;
> + u16 flags;
> + u16 reserved;
> +};
> +
> +/**
> + * struct acpi_nfit_idt - Interleave description Table
> + */
> +struct acpi_nfit_idt {
> + u16 type;
> + u16 length;
> + u16 idt_index;
> + u16 reserved;
> + u32 num_lines;
> + u32 line_size;
> + u32 line_offset[0];
> +};
> +
> +/**
> + * struct acpi_nfit_smbios - SMBIOS Management Information Table
> + */
> +struct acpi_nfit_smbios {
> + u16 type;
> + u16 length;
> + u32 reserved;
> + u8 data[0];
> +};
> +
> +/**
> + * struct acpi_nfit_dcr - NVDIMM Control Region Table
> + * @fic: Format Interface Code
> + * @cmd_offset: command registers relative to block control window
> + * @status_offset: status registers relative to block control window
> + */
> +struct acpi_nfit_dcr {
> + u16 type;
> + u16 length;
> + u16 dcr_index;
> + u16 vendor_id;
> + u16 device_id;
> + u16 revision_id;
> + u16 sub_vendor_id;
> + u16 sub_device_id;
> + u16 sub_revision_id;
> + u8 reserved[6];
> + u32 serial_number;
> + u16 fic;
> + u16 num_bcw;
> + u64 bcw_size;
> + u64 cmd_offset;
> + u64 cmd_size;
> + u64 status_offset;
> + u64 status_size;
> + u16 flags;
> + u8 reserved2[6];
> +};
> +
> +/**
> + * struct acpi_nfit_bdw - NVDIMM Block Data Window Region Table
> + */
> +struct acpi_nfit_bdw {
> + u16 type;
> + u16 length;
> + u16 dcr_index;
> + u16 num_bdw;
> + u64 bdw_offset;
> + u64 bdw_size;
> + u64 blk_capacity;
> + u64 blk_offset;
> +};
> +
> +/**
> + * struct acpi_nfit_flush - Flush Hint Address Structure
> + */
> +struct acpi_nfit_flush {
> + u16 type;
> + u16 length;
> + u32 nfit_handle;
> + u16 num_hints;
> + u8 reserved[6];
> + u64 hint_addr[0];
> +};
> +
> +struct nfit_spa {
> + struct acpi_nfit_spa *spa;
> + struct list_head list;
> +};
> +
> +struct nfit_dcr {
> + struct acpi_nfit_dcr *dcr;
> + struct list_head list;
> +};
> +
> +struct nfit_bdw {
> + struct acpi_nfit_bdw *bdw;
> + struct list_head list;
> +};
> +
> +struct nfit_memdev {
> + struct acpi_nfit_memdev *memdev;
> + struct list_head list;
> +};
> +
> +/* assembled tables for a given dimm/memory-device */
> +struct nfit_mem {
> + struct acpi_nfit_memdev *memdev_dcr;
> + struct acpi_nfit_memdev *memdev_pmem;
> + struct acpi_nfit_dcr *dcr;
> + struct acpi_nfit_bdw *bdw;
> + struct acpi_nfit_spa *spa_dcr;
> + struct acpi_nfit_spa *spa_bdw;
> + struct list_head list;
> +};
> +
> +struct acpi_nfit_desc {
> + struct nd_bus_descriptor nd_desc;
> + struct acpi_nfit *nfit;
> + struct list_head memdevs;
> + struct list_head dimms;
> + struct list_head spas;
> + struct list_head dcrs;
> + struct list_head bdws;
> + struct nd_bus *nd_bus;
> + struct device *dev;
> +};
> +
> +static inline struct acpi_nfit_memdev *__to_nfit_memdev(struct nfit_mem *nfit_mem)
> +{
> + if (nfit_mem->memdev_dcr)
> + return nfit_mem->memdev_dcr;
> + return nfit_mem->memdev_pmem;
> +}
> +#endif /* __NFIT_H__ */
> diff --git a/drivers/block/nd/core.c b/drivers/block/nd/core.c
> new file mode 100644
> index 000000000000..3cccdbc0f3b7
> --- /dev/null
> +++ b/drivers/block/nd/core.c
> @@ -0,0 +1,67 @@
> +/*
> + * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of version 2 of the GNU General Public License as
> + * published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope that it will be useful, but
> + * WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> + * General Public License for more details.
> + */
> +#include <linux/export.h>
> +#include <linux/module.h>
> +#include <linux/device.h>
> +#include <linux/slab.h>
> +#include "nd-private.h"
> +#include "libnd.h"
> +
> +static DEFINE_IDA(nd_ida);
> +
> +static void nd_bus_release(struct device *dev)
> +{
> + struct nd_bus *nd_bus = container_of(dev, struct nd_bus, dev);
> +
> + ida_simple_remove(&nd_ida, nd_bus->id);
> + kfree(nd_bus);
> +}
> +
> +struct nd_bus *nd_bus_register(struct device *parent,
> + struct nd_bus_descriptor *nd_desc)
> +{
> + struct nd_bus *nd_bus = kzalloc(sizeof(*nd_bus), GFP_KERNEL);
> + int rc;
> +
> + if (!nd_bus)
> + return NULL;
> + nd_bus->id = ida_simple_get(&nd_ida, 0, 0, GFP_KERNEL);
> + if (nd_bus->id < 0) {
> + kfree(nd_bus);
> + return NULL;
> + }
> + nd_bus->nd_desc = nd_desc;
> + nd_bus->dev.parent = parent;
> + nd_bus->dev.release = nd_bus_release;
> + dev_set_name(&nd_bus->dev, "ndbus%d", nd_bus->id);
> + rc = device_register(&nd_bus->dev);
> + if (rc) {
> + dev_dbg(&nd_bus->dev, "device registration failed: %d\n", rc);
> + put_device(&nd_bus->dev);
> + return NULL;
> + }
> +
> + return nd_bus;
> +}
> +EXPORT_SYMBOL_GPL(nd_bus_register);
> +
> +void nd_bus_unregister(struct nd_bus *nd_bus)
> +{
> + if (!nd_bus)
> + return;
> + device_unregister(&nd_bus->dev);
> +}
> +EXPORT_SYMBOL_GPL(nd_bus_unregister);
> +
> +MODULE_LICENSE("GPL v2");
> +MODULE_AUTHOR("Intel Corporation");
> diff --git a/drivers/block/nd/libnd.h b/drivers/block/nd/libnd.h
> new file mode 100644
> index 000000000000..163832937e9c
> --- /dev/null
> +++ b/drivers/block/nd/libnd.h
> @@ -0,0 +1,33 @@
> +/*
> + * libnd - Non-volatile-memory Devices Subsystem
> + *
> + * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of version 2 of the GNU General Public License as
> + * published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope that it will be useful, but
> + * WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> + * General Public License for more details.
> + */
> +#ifndef __LIBND_H__
> +#define __LIBND_H__
> +struct nd_dimm;
> +struct nd_bus_descriptor;
> +typedef int (*ndctl_fn)(struct nd_bus_descriptor *nd_desc,
> + struct nd_dimm *nd_dimm, unsigned int cmd, void *buf,
> + unsigned int buf_len);
> +
> +struct nd_bus_descriptor {
> + unsigned long dsm_mask;
> + char *provider_name;
> + ndctl_fn ndctl;
> +};
> +
> +struct nd_bus;
> +struct nd_bus *nd_bus_register(struct device *parent,
> + struct nd_bus_descriptor *nfit_desc);
> +void nd_bus_unregister(struct nd_bus *nd_bus);
> +#endif /* __LIBND_H__ */
> diff --git a/drivers/block/nd/nd-private.h b/drivers/block/nd/nd-private.h
> new file mode 100644
> index 000000000000..3dbab29fa0f9
> --- /dev/null
> +++ b/drivers/block/nd/nd-private.h
> @@ -0,0 +1,23 @@
> +/*
> + * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of version 2 of the GNU General Public License as
> + * published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope that it will be useful, but
> + * WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> + * General Public License for more details.
> + */
> +#ifndef __ND_PRIVATE_H__
> +#define __ND_PRIVATE_H__
> +#include <linux/device.h>
> +#include "libnd.h"
> +
> +struct nd_bus {
> + struct nd_bus_descriptor *nd_desc;
> + struct device dev;
> + int id;
> +};
> +#endif /* __ND_PRIVATE_H__ */
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
--
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.
On Thu, Apr 30, 2015 at 4:23 PM, Rafael J. Wysocki <[email protected]> wrote:
> On Tuesday, April 28, 2015 02:24:23 PM Dan Williams wrote:
>> 1/ Autodetect an NFIT table for the ACPI namespace device with _HID of
>> "ACPI0012"
>>
>> 2/ libnd bus registration
>>
>> The NFIT provided by ACPI is one possible method by which platforms will
>> discover NVDIMM resources. However, the intent of the nd_bus_descriptor
>> abstraction is to abstract "provider" specific details, leaving libnd
>> to be independent of the specific NVDIMM resource discovery mechanism.
>> This flexibility is later exploited later to implement custom-defined nd
>> buses.
>>
>> Cc: <[email protected]>
>> Cc: Robert Moore <[email protected]>
>> Cc: Rafael J. Wysocki <[email protected]>
>> Signed-off-by: Dan Williams <[email protected]>
>> ---
>> drivers/block/Kconfig | 2
>> drivers/block/Makefile | 1
>> drivers/block/nd/Kconfig | 40 +++
>> drivers/block/nd/Makefile | 6 +
>> drivers/block/nd/acpi.c | 475 +++++++++++++++++++++++++++++++++++++++++
>> drivers/block/nd/acpi_nfit.h | 254 ++++++++++++++++++++++
>> drivers/block/nd/core.c | 67 ++++++
>> drivers/block/nd/libnd.h | 33 +++
>> drivers/block/nd/nd-private.h | 23 ++
>> 9 files changed, 901 insertions(+)
>> create mode 100644 drivers/block/nd/Kconfig
>> create mode 100644 drivers/block/nd/Makefile
>> create mode 100644 drivers/block/nd/acpi.c
>> create mode 100644 drivers/block/nd/acpi_nfit.h
>> create mode 100644 drivers/block/nd/core.c
>> create mode 100644 drivers/block/nd/libnd.h
>> create mode 100644 drivers/block/nd/nd-private.h
>>
>> diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig
>> index eb1fed5bd516..dfe40e5ca9bd 100644
>> --- a/drivers/block/Kconfig
>> +++ b/drivers/block/Kconfig
>> @@ -321,6 +321,8 @@ config BLK_DEV_NVME
>> To compile this driver as a module, choose M here: the
>> module will be called nvme.
>>
>> +source "drivers/block/nd/Kconfig"
>> +
>> config BLK_DEV_SKD
>> tristate "STEC S1120 Block Driver"
>> depends on PCI
>> diff --git a/drivers/block/Makefile b/drivers/block/Makefile
>> index 9cc6c18a1c7e..07a6acecf4d8 100644
>> --- a/drivers/block/Makefile
>> +++ b/drivers/block/Makefile
>> @@ -24,6 +24,7 @@ obj-$(CONFIG_CDROM_PKTCDVD) += pktcdvd.o
>> obj-$(CONFIG_MG_DISK) += mg_disk.o
>> obj-$(CONFIG_SUNVDC) += sunvdc.o
>> obj-$(CONFIG_BLK_DEV_NVME) += nvme.o
>> +obj-$(CONFIG_ND_DEVICES) += nd/
>> obj-$(CONFIG_BLK_DEV_SKD) += skd.o
>> obj-$(CONFIG_BLK_DEV_OSD) += osdblk.o
>>
>> diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
>> new file mode 100644
>> index 000000000000..6d5d6b732f82
>> --- /dev/null
>> +++ b/drivers/block/nd/Kconfig
>> @@ -0,0 +1,40 @@
>> +menuconfig ND_DEVICES
>> + bool "NVDIMM Support"
>> + depends on PHYS_ADDR_T_64BIT
>> + help
>> + Generic support for non-volatile memory devices including
>> + ACPI-6-NFIT defined resources. On platforms that define an
>> + NFIT, or otherwise can discover NVDIMM resources, a libnd
>> + bus is registered to advertise PMEM (persistent memory)
>> + namespaces (/dev/pmemX) and BLK (sliding mmio window(s))
>> + namespaces (/dev/ndX). A PMEM namespace refers to a memory
>> + resource that may span multiple DIMMs and support DAX (see
>> + CONFIG_DAX). A BLK namespace refers to an NVDIMM control
>> + region which exposes an mmio register set for windowed
>> + access mode to non-volatile memory.
>> +
>> +if ND_DEVICES
>> +
>> +config LIBND
>> + tristate "LIBND: libnd device driver support"
>> + help
>> + Platform agnostic device model for a libnd bus. Publishes
>> + resources for a PMEM (persistent-memory) driver and/or BLK
>> + (sliding mmio window(s)) driver to attach. Exposes a device
>> + topology under a "ndX" bus device, a "/dev/ndctlX" bus-ioctl
>> + message passing interface, and a "/dev/nmemX" dimm-ioctl
>> + message interface for each memory device registered on the
>> + bus. instance. A userspace library "ndctl" provides an API
>> + to enumerate/manage this subsystem.
>> +
>> +config ND_ACPI
>> + tristate "ACPI: NFIT to libnd bus support"
>> + select LIBND
>> + depends on ACPI
>> + help
>> + Infrastructure to probe ACPI 6 compliant platforms for
>> + NVDIMMs (NFIT) and register a libnd device tree. In
>> + addition to storage devices this also enables libnd craft
>> + ACPI._DSM messages for platform/dimm configuration.
>
> I'm wondering if the two CONFIG options above really need to be user-selectable?
>
> For example, what reason people (who've already selected ND_DEVICES) may have
> for not selecting ND_ACPI if ACPI is set?
Later on in the series we introduce ND_E820 which supports creating a
libnd-bus from e820-type-12 memory ranges on pre-NFIT systems. I'm
also considering a configfs defined libnd-bus because e820 types are
not nearly enough information to safely define nvdimm resources
outside of NFIT.
>> +
>> +endif
>> diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
>> new file mode 100644
>> index 000000000000..944b5947c0cb
>> --- /dev/null
>> +++ b/drivers/block/nd/Makefile
>> @@ -0,0 +1,6 @@
>> +obj-$(CONFIG_LIBND) += libnd.o
>> +obj-$(CONFIG_ND_ACPI) += nd_acpi.o
>> +
>> +nd_acpi-y := acpi.o
>> +
>> +libnd-y := core.o
>
> OK, so it looks like no modules, just built-in code, right?
>
Um, no, both CONFIG_ND_ACPI and CONFIG_LIBND can be =m.
>> diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
>> new file mode 100644
>> index 000000000000..9f0b24390d1b
>> --- /dev/null
>> +++ b/drivers/block/nd/acpi.c
>> @@ -0,0 +1,475 @@
>> +/*
>> + * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
>> + *
>> + * This program is free software; you can redistribute it and/or modify
>> + * it under the terms of version 2 of the GNU General Public License as
>> + * published by the Free Software Foundation.
>> + *
>> + * This program is distributed in the hope that it will be useful, but
>> + * WITHOUT ANY WARRANTY; without even the implied warranty of
>> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
>> + * General Public License for more details.
>> + */
>> +#include <linux/list_sort.h>
>> +#include <linux/module.h>
>> +#include <linux/list.h>
>> +#include <linux/acpi.h>
>> +#include "acpi_nfit.h"
>> +#include "libnd.h"
>> +
>> +static bool warn_checksum;
>> +module_param(warn_checksum, bool, S_IRUGO|S_IWUSR);
>> +MODULE_PARM_DESC(warn_checksum, "Turn checksum errors into warnings");
>> +
>> +enum {
>> + NFIT_ACPI_NOTIFY_TABLE = 0x80,
>> +};
>> +
>> +static int nd_acpi_ctl(struct nd_bus_descriptor *nd_desc,
>> + struct nd_dimm *nd_dimm, unsigned int cmd, void *buf,
>> + unsigned int buf_len)
>> +{
>> + return -ENOTTY;
>> +}
>
> Why -ENOTTY? And why not to leave a NULL entry for this instead and
> make the library fail it in that case?
Yes, I should have deferred this to the patch that fills in
nd_acpi_ctl later in the series. Having this stub here in this patch
just trips up reviewers, sorry.
>> +static const char *spa_type_name(u16 type)
>> +{
>> + switch (type) {
>> + case NFIT_SPA_VOLATILE: return "volatile";
>> + case NFIT_SPA_PM: return "pmem";
>> + case NFIT_SPA_DCR: return "dimm-control-region";
>> + case NFIT_SPA_BDW: return "block-data-window";
>> + default: return "unknown";
>> + }
>> +}
>> +
>> +static int nfit_spa_type(struct acpi_nfit_spa *spa)
>> +{
>> + if (memcmp(&nfit_spa_uuid_volatile, spa->type_uuid, 16) == 0)
>> + return NFIT_SPA_VOLATILE;
>> +
>> + if (memcmp(&nfit_spa_uuid_pm, spa->type_uuid, 16) == 0)
>> + return NFIT_SPA_PM;
>> +
>> + if (memcmp(&nfit_spa_uuid_dcr, spa->type_uuid, 16) == 0)
>> + return NFIT_SPA_DCR;
>> +
>> + if (memcmp(&nfit_spa_uuid_bdw, spa->type_uuid, 16) == 0)
>> + return NFIT_SPA_BDW;
>> +
>> + if (memcmp(&nfit_spa_uuid_vdisk, spa->type_uuid, 16) == 0)
>> + return NFIT_SPA_VDISK;
>> +
>> + if (memcmp(&nfit_spa_uuid_vcd, spa->type_uuid, 16) == 0)
>> + return NFIT_SPA_VCD;
>> +
>> + if (memcmp(&nfit_spa_uuid_pdisk, spa->type_uuid, 16) == 0)
>> + return NFIT_SPA_PDISK;
>> +
>> + if (memcmp(&nfit_spa_uuid_pcd, spa->type_uuid, 16) == 0)
>> + return NFIT_SPA_PCD;
>> +
>> + return -1;
>> +}
>> +
>> +struct nfit_table_header {
>> + __le16 type;
>> + __le16 length;
>> +};
>
> That you'll be able to get from ACPICA I suppose?
Yes.
>> +
>> +static void *add_table(struct acpi_nfit_desc *acpi_desc, void *table, const void *end)
>> +{
>> + struct device *dev = acpi_desc->dev;
>> + struct nfit_table_header *hdr;
>> + void *err = ERR_PTR(-ENOMEM);
>> +
>> + if (table >= end)
>> + return NULL;
>> +
>> + hdr = (struct nfit_table_header *) table;
>> + switch (hdr->type) {
>> + case NFIT_TABLE_SPA: {
>> + struct nfit_spa *nfit_spa = devm_kzalloc(dev, sizeof(*nfit_spa),
>> + GFP_KERNEL);
>> + struct acpi_nfit_spa *spa = table;
>> +
>> + if (!nfit_spa)
>> + return err;
>> + INIT_LIST_HEAD(&nfit_spa->list);
>> + nfit_spa->spa = spa;
>> + list_add_tail(&nfit_spa->list, &acpi_desc->spas);
>> + dev_dbg(dev, "%s: spa index: %d type: %s\n", __func__,
>> + spa->spa_index,
>> + spa_type_name(nfit_spa_type(spa)));
>> + break;
>> + }
>> + case NFIT_TABLE_MEM: {
>> + struct nfit_memdev *nfit_memdev = devm_kzalloc(dev,
>> + sizeof(*nfit_memdev), GFP_KERNEL);
>> + struct acpi_nfit_memdev *memdev = table;
>> +
>> + if (!nfit_memdev)
>> + return err;
>> + INIT_LIST_HEAD(&nfit_memdev->list);
>> + nfit_memdev->memdev = memdev;
>> + list_add_tail(&nfit_memdev->list, &acpi_desc->memdevs);
>> + dev_dbg(dev, "%s: memdev handle: %#x spa: %d dcr: %d\n",
>> + __func__, memdev->nfit_handle, memdev->spa_index,
>> + memdev->dcr_index);
>> + break;
>> + }
>> + case NFIT_TABLE_DCR: {
>> + struct nfit_dcr *nfit_dcr = devm_kzalloc(dev, sizeof(*nfit_dcr),
>> + GFP_KERNEL);
>> + struct acpi_nfit_dcr *dcr = table;
>> +
>> + if (!nfit_dcr)
>> + return err;
>> + INIT_LIST_HEAD(&nfit_dcr->list);
>> + nfit_dcr->dcr = dcr;
>> + list_add_tail(&nfit_dcr->list, &acpi_desc->dcrs);
>> + dev_dbg(dev, "%s: dcr index: %d num_bcw: %d\n", __func__,
>> + dcr->dcr_index, dcr->num_bcw);
>> + break;
>> + }
>> + case NFIT_TABLE_BDW: {
>> + struct nfit_bdw *nfit_bdw = devm_kzalloc(dev, sizeof(*nfit_bdw),
>> + GFP_KERNEL);
>> + struct acpi_nfit_bdw *bdw = table;
>> +
>> + if (!nfit_bdw)
>> + return err;
>> + INIT_LIST_HEAD(&nfit_bdw->list);
>> + nfit_bdw->bdw = bdw;
>> + list_add_tail(&nfit_bdw->list, &acpi_desc->bdws);
>> + dev_dbg(dev, "%s: bdw dcr: %d num_bdw: %d\n", __func__,
>> + bdw->dcr_index, bdw->num_bdw);
>> + break;
>> + }
>> + /* TODO */
>> + case NFIT_TABLE_IDT:
>> + dev_dbg(dev, "%s: idt\n", __func__);
>> + break;
>> + case NFIT_TABLE_FLUSH:
>> + dev_dbg(dev, "%s: flush\n", __func__);
>> + break;
>> + case NFIT_TABLE_SMBIOS:
>> + dev_dbg(dev, "%s: smbios\n", __func__);
>> + break;
>> + default:
>> + dev_err(dev, "unknown table '%d' parsing nfit\n", hdr->type);
>> + return ERR_PTR(-ENXIO);
>> + }
>> +
>> + return table + hdr->length;
>> +}
>> +
>> +static void nfit_mem_find_spa_bdw(struct acpi_nfit_desc *acpi_desc,
>> + struct nfit_mem *nfit_mem)
>> +{
>> + u32 nfit_handle = __to_nfit_memdev(nfit_mem)->nfit_handle;
>> + u16 dcr_index = nfit_mem->dcr->dcr_index;
>> + struct nfit_spa *nfit_spa;
>> +
>> + list_for_each_entry(nfit_spa, &acpi_desc->spas, list) {
>> + u16 spa_index = nfit_spa->spa->spa_index;
>> + int type = nfit_spa_type(nfit_spa->spa);
>> + struct nfit_memdev *nfit_memdev;
>> +
>> + if (type != NFIT_SPA_BDW)
>> + continue;
>> +
>> + list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) {
>> + if (nfit_memdev->memdev->spa_index != spa_index)
>> + continue;
>> + if (nfit_memdev->memdev->nfit_handle != nfit_handle)
>> + continue;
>> + if (nfit_memdev->memdev->dcr_index != dcr_index)
>> + continue;
>> +
>> + nfit_mem->spa_bdw = nfit_spa->spa;
>> + return;
>> + }
>> + }
>> +
>> + dev_dbg(acpi_desc->dev, "SPA-BDW not found for SPA-DCR %d\n",
>> + nfit_mem->spa_dcr->spa_index);
>> + nfit_mem->bdw = NULL;
>> +}
>> +
>> +static int nfit_mem_add(struct acpi_nfit_desc *acpi_desc,
>> + struct nfit_mem *nfit_mem, struct acpi_nfit_spa *spa)
>> +{
>> + u16 dcr_index = __to_nfit_memdev(nfit_mem)->dcr_index;
>> + struct nfit_dcr *nfit_dcr;
>> + struct nfit_bdw *nfit_bdw;
>> +
>> + list_for_each_entry(nfit_dcr, &acpi_desc->dcrs, list) {
>> + if (nfit_dcr->dcr->dcr_index != dcr_index)
>> + continue;
>> + nfit_mem->dcr = nfit_dcr->dcr;
>> + break;
>> + }
>> +
>> + if (!nfit_mem->dcr) {
>> + dev_dbg(acpi_desc->dev, "SPA %d missing:%s%s\n", spa->spa_index,
>> + __to_nfit_memdev(nfit_mem) ? "" : " MEMDEV",
>> + nfit_mem->dcr ? "" : " DCR");
>> + return -ENODEV;
>> + }
>> +
>> + /*
>> + * We've found enough to create an nd_dimm, optionally
>> + * find an associated BDW
>> + */
>> + list_add(&nfit_mem->list, &acpi_desc->dimms);
>> +
>> + list_for_each_entry(nfit_bdw, &acpi_desc->bdws, list) {
>> + if (nfit_bdw->bdw->dcr_index != dcr_index)
>> + continue;
>> + nfit_mem->bdw = nfit_bdw->bdw;
>> + break;
>> + }
>> +
>> + if (!nfit_mem->bdw)
>> + return 0;
>> +
>> + nfit_mem_find_spa_bdw(acpi_desc, nfit_mem);
>> + return 0;
>> +}
>> +
>> +static int nfit_mem_dcr_init(struct acpi_nfit_desc *acpi_desc,
>> + struct acpi_nfit_spa *spa)
>> +{
>> + struct nfit_mem *nfit_mem, *found;
>> + struct nfit_memdev *nfit_memdev;
>> + int type = nfit_spa_type(spa);
>> + u16 dcr_index;
>> +
>> + switch (type) {
>> + case NFIT_SPA_DCR:
>> + case NFIT_SPA_PM:
>> + break;
>> + default:
>> + return 0;
>> + }
>> +
>> + list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) {
>> + int rc;
>> +
>> + if (nfit_memdev->memdev->spa_index != spa->spa_index)
>> + continue;
>> + found = NULL;
>> + dcr_index = nfit_memdev->memdev->dcr_index;
>> + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list)
>> + if (__to_nfit_memdev(nfit_mem)->dcr_index == dcr_index) {
>> + found = nfit_mem;
>> + break;
>> + }
>> +
>> + if (found)
>> + nfit_mem = found;
>> + else {
>> + nfit_mem = devm_kzalloc(acpi_desc->dev,
>> + sizeof(*nfit_mem), GFP_KERNEL);
>> + if (!nfit_mem)
>> + return -ENOMEM;
>> + INIT_LIST_HEAD(&nfit_mem->list);
>> + }
>> +
>> + if (type == NFIT_SPA_DCR) {
>> + /* multiple dimms may share a SPA when interleaved */
>> + nfit_mem->spa_dcr = spa;
>> + nfit_mem->memdev_dcr = nfit_memdev->memdev;
>> + } else {
>> + /*
>> + * A single dimm may belong to multiple SPA-PM
>> + * ranges, record at least one in addition to
>> + * any SPA-DCR range.
>> + */
>> + nfit_mem->memdev_pmem = nfit_memdev->memdev;
>> + }
>> +
>> + if (found)
>> + continue;
>> +
>> + rc = nfit_mem_add(acpi_desc, nfit_mem, spa);
>> + if (rc)
>> + return rc;
>> + }
>> +
>> + return 0;
>> +}
>> +
>> +static int nfit_mem_cmp(void *priv, struct list_head *__a, struct list_head *__b)
>> +{
>> + struct nfit_mem *a = container_of(__a, typeof(*a), list);
>> + struct nfit_mem *b = container_of(__b, typeof(*b), list);
>> + u32 handleA, handleB;
>> +
>> + handleA = __to_nfit_memdev(a)->nfit_handle;
>> + handleB = __to_nfit_memdev(b)->nfit_handle;
>> + if (handleA < handleB)
>> + return -1;
>> + else if (handleA > handleB)
>> + return 1;
>> + return 0;
>> +}
>> +
>> +static int nfit_mem_init(struct acpi_nfit_desc *acpi_desc)
>> +{
>> + struct nfit_spa *nfit_spa;
>> +
>> + /*
>> + * For each SPA-DCR or SPA-PMEM address range find its
>> + * corresponding MEMDEV(s). From each MEMDEV find the
>> + * corresponding DCR. Then, if we're operating on a SPA-DCR,
>> + * try to find a SPA-BDW and a corresponding BDW that references
>> + * the DCR. Throw it all into an nfit_mem object. Note, that
>> + * BDWs are optional.
>> + */
>> + list_for_each_entry(nfit_spa, &acpi_desc->spas, list) {
>> + int rc;
>> +
>> + rc = nfit_mem_dcr_init(acpi_desc, nfit_spa->spa);
>> + if (rc)
>> + return rc;
>> + }
>> +
>> + list_sort(NULL, &acpi_desc->dimms, nfit_mem_cmp);
>> +
>> + return 0;
>> +}
>> +
>> +static int nd_acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, acpi_size sz)
>> +{
>> + struct device *dev = acpi_desc->dev;
>> + const void *end;
>> + u8 *data, sum;
>> + acpi_size i;
>> +
>> + INIT_LIST_HEAD(&acpi_desc->spas);
>> + INIT_LIST_HEAD(&acpi_desc->dcrs);
>> + INIT_LIST_HEAD(&acpi_desc->bdws);
>> + INIT_LIST_HEAD(&acpi_desc->memdevs);
>> + INIT_LIST_HEAD(&acpi_desc->dimms);
>> +
>> + data = (u8 *) acpi_desc->nfit;
>> + for (i = 0, sum = 0; i < sz; i++)
>> + sum += readb(data + i);
>> + if (sum != 0 && !warn_checksum) {
>> + dev_dbg(dev, "%s: nfit checksum failure\n", __func__);
>> + return -ENXIO;
>> + }
>> + WARN_TAINT_ONCE(sum != 0, TAINT_FIRMWARE_WORKAROUND,
>> + "nfit checksum failure, continuing...\n");
>> +
>> + end = data + sz;
>> + data += sizeof(struct acpi_nfit);
>> + while (!IS_ERR_OR_NULL(data))
>> + data = add_table(acpi_desc, data, end);
>
> This looks like we are expecting a series of tables here and we're going to
> fail the whole discovery if just one of them in invalid.
>
> I wonder if it would be practical to skip just the invalid ones instead?
Yes, for future-proofing we should just ignore tables that we don't
understand. Will fix.
>
>> +
>> + if (IS_ERR(data)) {
>> + dev_dbg(dev, "%s: nfit table parsing error: %ld\n", __func__,
>> + PTR_ERR(data));
>> + return PTR_ERR(data);
>> + }
>> +
>> + if (nfit_mem_init(acpi_desc) != 0)
>> + return -ENOMEM;
>> +
>> + return 0;
>> +}
>> +
>> +static int nd_acpi_add(struct acpi_device *adev)
>> +{
>> + struct nd_bus_descriptor *nd_desc;
>> + struct acpi_nfit_desc *acpi_desc;
>> + struct device *dev = &adev->dev;
>> + struct acpi_table_header *tbl;
>> + acpi_status status = AE_OK;
>> + acpi_size sz;
>> + int rc;
>> +
>> + status = acpi_get_table_with_size("NFIT", 0, &tbl, &sz);
>> + if (ACPI_FAILURE(status)) {
>> + dev_err(dev, "failed to find NFIT\n");
>> + return -ENXIO;
>> + }
>> +
>> + acpi_desc = devm_kzalloc(dev, sizeof(*acpi_desc), GFP_KERNEL);
>> + if (!acpi_desc)
>> + return -ENOMEM;
>> +
>> + dev_set_drvdata(dev, acpi_desc);
>> + acpi_desc->dev = dev;
>> + acpi_desc->nfit = (struct acpi_nfit *) tbl;
>> + nd_desc = &acpi_desc->nd_desc;
>> + nd_desc->provider_name = "ACPI.NFIT";
>> + nd_desc->ndctl = nd_acpi_ctl;
>> +
>> + acpi_desc->nd_bus = nd_bus_register(dev, nd_desc);
>> + if (!acpi_desc->nd_bus)
>> + return -ENXIO;
>> +
>> + rc = nd_acpi_nfit_init(acpi_desc, sz);
>> + if (rc) {
>> + nd_bus_unregister(acpi_desc->nd_bus);
>> + return rc;
>> + }
>> + return 0;
>> +}
>> +
>> +static int nd_acpi_remove(struct acpi_device *adev)
>> +{
>> + struct acpi_nfit_desc *acpi_desc = dev_get_drvdata(&adev->dev);
>> +
>> + nd_bus_unregister(acpi_desc->nd_bus);
>> + return 0;
>> +}
>> +
>> +static void nd_acpi_notify(struct acpi_device *adev, u32 event)
>> +{
>> + /* TODO: handle ACPI_NOTIFY_BUS_CHECK notification */
>> + dev_dbg(&adev->dev, "%s: event: %d\n", __func__, event);
>> +}
>> +
>> +static const struct acpi_device_id nd_acpi_ids[] = {
>> + { "ACPI0012", 0 },
>> + { "", 0 },
>> +};
>> +MODULE_DEVICE_TABLE(acpi, nd_acpi_ids);
>> +
>> +static struct acpi_driver nd_acpi_driver = {
>> + .name = KBUILD_MODNAME,
>> + .ids = nd_acpi_ids,
>> + .flags = ACPI_DRIVER_ALL_NOTIFY_EVENTS,
>> + .ops = {
>> + .add = nd_acpi_add,
>> + .remove = nd_acpi_remove,
>> + .notify = nd_acpi_notify
>> + },
>> +};
>
> Since this is going to be non-modular built-in code, please use an ACPI
> scan handler instead of using a driver here. acpi_memhotplug.c does that,
> you can use it as an example, but I guess you don't need to enable hotplug
> for it to start with.
No, you misunderstood, this will certainly be modular and loaded on-demand.
>
>> +
>> +static __init int nd_acpi_init(void)
>> +{
>> + BUILD_BUG_ON(sizeof(struct acpi_nfit) != 40);
>> + BUILD_BUG_ON(sizeof(struct acpi_nfit_spa) != 56);
>> + BUILD_BUG_ON(sizeof(struct acpi_nfit_memdev) != 48);
>> + BUILD_BUG_ON(sizeof(struct acpi_nfit_idt) != 16);
>> + BUILD_BUG_ON(sizeof(struct acpi_nfit_smbios) != 8);
>> + BUILD_BUG_ON(sizeof(struct acpi_nfit_dcr) != 80);
>> + BUILD_BUG_ON(sizeof(struct acpi_nfit_bdw) != 40);
>> +
>> + return acpi_bus_register_driver(&nd_acpi_driver);
>> +}
>> +
>> +static __exit void nd_acpi_exit(void)
>> +{
>> + acpi_bus_unregister_driver(&nd_acpi_driver);
>> +}
>> +
>> +module_init(nd_acpi_init);
>> +module_exit(nd_acpi_exit);
>> +MODULE_LICENSE("GPL v2");
>> +MODULE_AUTHOR("Intel Corporation");
>> diff --git a/drivers/block/nd/acpi_nfit.h b/drivers/block/nd/acpi_nfit.h
>> new file mode 100644
>> index 000000000000..e0b0f12736bf
>> --- /dev/null
>> +++ b/drivers/block/nd/acpi_nfit.h
>
> I'm assuming that the below is coordinated with Bob and David and will be
> changed to use ACPICA-provided definitions going forward.
>
> Is that correct?
Yes, as soon as definitions those are available we will drop this
header and rebase on the ACPICA implementation.
[..]
On Thursday, April 30, 2015 05:39:06 PM Dan Williams wrote:
> On Thu, Apr 30, 2015 at 4:23 PM, Rafael J. Wysocki <[email protected]> wrote:
> > On Tuesday, April 28, 2015 02:24:23 PM Dan Williams wrote:
> >> 1/ Autodetect an NFIT table for the ACPI namespace device with _HID of
> >> "ACPI0012"
> >>
> >> 2/ libnd bus registration
> >>
> >> The NFIT provided by ACPI is one possible method by which platforms will
> >> discover NVDIMM resources. However, the intent of the nd_bus_descriptor
> >> abstraction is to abstract "provider" specific details, leaving libnd
> >> to be independent of the specific NVDIMM resource discovery mechanism.
> >> This flexibility is later exploited later to implement custom-defined nd
> >> buses.
> >>
> >> Cc: <[email protected]>
> >> Cc: Robert Moore <[email protected]>
> >> Cc: Rafael J. Wysocki <[email protected]>
> >> Signed-off-by: Dan Williams <[email protected]>
> >> ---
> >> drivers/block/Kconfig | 2
> >> drivers/block/Makefile | 1
> >> drivers/block/nd/Kconfig | 40 +++
> >> drivers/block/nd/Makefile | 6 +
> >> drivers/block/nd/acpi.c | 475 +++++++++++++++++++++++++++++++++++++++++
> >> drivers/block/nd/acpi_nfit.h | 254 ++++++++++++++++++++++
> >> drivers/block/nd/core.c | 67 ++++++
> >> drivers/block/nd/libnd.h | 33 +++
> >> drivers/block/nd/nd-private.h | 23 ++
> >> 9 files changed, 901 insertions(+)
> >> create mode 100644 drivers/block/nd/Kconfig
> >> create mode 100644 drivers/block/nd/Makefile
> >> create mode 100644 drivers/block/nd/acpi.c
> >> create mode 100644 drivers/block/nd/acpi_nfit.h
> >> create mode 100644 drivers/block/nd/core.c
> >> create mode 100644 drivers/block/nd/libnd.h
> >> create mode 100644 drivers/block/nd/nd-private.h
> >>
> >> diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig
> >> index eb1fed5bd516..dfe40e5ca9bd 100644
> >> --- a/drivers/block/Kconfig
> >> +++ b/drivers/block/Kconfig
> >> @@ -321,6 +321,8 @@ config BLK_DEV_NVME
> >> To compile this driver as a module, choose M here: the
> >> module will be called nvme.
> >>
> >> +source "drivers/block/nd/Kconfig"
> >> +
> >> config BLK_DEV_SKD
> >> tristate "STEC S1120 Block Driver"
> >> depends on PCI
> >> diff --git a/drivers/block/Makefile b/drivers/block/Makefile
> >> index 9cc6c18a1c7e..07a6acecf4d8 100644
> >> --- a/drivers/block/Makefile
> >> +++ b/drivers/block/Makefile
> >> @@ -24,6 +24,7 @@ obj-$(CONFIG_CDROM_PKTCDVD) += pktcdvd.o
> >> obj-$(CONFIG_MG_DISK) += mg_disk.o
> >> obj-$(CONFIG_SUNVDC) += sunvdc.o
> >> obj-$(CONFIG_BLK_DEV_NVME) += nvme.o
> >> +obj-$(CONFIG_ND_DEVICES) += nd/
> >> obj-$(CONFIG_BLK_DEV_SKD) += skd.o
> >> obj-$(CONFIG_BLK_DEV_OSD) += osdblk.o
> >>
> >> diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig
> >> new file mode 100644
> >> index 000000000000..6d5d6b732f82
> >> --- /dev/null
> >> +++ b/drivers/block/nd/Kconfig
> >> @@ -0,0 +1,40 @@
> >> +menuconfig ND_DEVICES
> >> + bool "NVDIMM Support"
> >> + depends on PHYS_ADDR_T_64BIT
> >> + help
> >> + Generic support for non-volatile memory devices including
> >> + ACPI-6-NFIT defined resources. On platforms that define an
> >> + NFIT, or otherwise can discover NVDIMM resources, a libnd
> >> + bus is registered to advertise PMEM (persistent memory)
> >> + namespaces (/dev/pmemX) and BLK (sliding mmio window(s))
> >> + namespaces (/dev/ndX). A PMEM namespace refers to a memory
> >> + resource that may span multiple DIMMs and support DAX (see
> >> + CONFIG_DAX). A BLK namespace refers to an NVDIMM control
> >> + region which exposes an mmio register set for windowed
> >> + access mode to non-volatile memory.
> >> +
> >> +if ND_DEVICES
> >> +
> >> +config LIBND
> >> + tristate "LIBND: libnd device driver support"
> >> + help
> >> + Platform agnostic device model for a libnd bus. Publishes
> >> + resources for a PMEM (persistent-memory) driver and/or BLK
> >> + (sliding mmio window(s)) driver to attach. Exposes a device
> >> + topology under a "ndX" bus device, a "/dev/ndctlX" bus-ioctl
> >> + message passing interface, and a "/dev/nmemX" dimm-ioctl
> >> + message interface for each memory device registered on the
> >> + bus. instance. A userspace library "ndctl" provides an API
> >> + to enumerate/manage this subsystem.
> >> +
> >> +config ND_ACPI
> >> + tristate "ACPI: NFIT to libnd bus support"
> >> + select LIBND
> >> + depends on ACPI
> >> + help
> >> + Infrastructure to probe ACPI 6 compliant platforms for
> >> + NVDIMMs (NFIT) and register a libnd device tree. In
> >> + addition to storage devices this also enables libnd craft
> >> + ACPI._DSM messages for platform/dimm configuration.
> >
> > I'm wondering if the two CONFIG options above really need to be user-selectable?
> >
> > For example, what reason people (who've already selected ND_DEVICES) may have
> > for not selecting ND_ACPI if ACPI is set?
>
>
> Later on in the series we introduce ND_E820 which supports creating a
> libnd-bus from e820-type-12 memory ranges on pre-NFIT systems. I'm
> also considering a configfs defined libnd-bus because e820 types are
> not nearly enough information to safely define nvdimm resources
> outside of NFIT.
I hope these are not mutually exclusive with ND_ACPI? Otherwise distros
will have problems with supporting them in one kernel.
If ND_E820 and ND_ACPI aren't mutually exclusive, I still don't see a good
enough reason for asking users about ND_ACPI. Why would I ever say "No"
here if I said "Yes" or "Module" to ND_DEVICES?
> >> +
> >> +endif
> >> diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
> >> new file mode 100644
> >> index 000000000000..944b5947c0cb
> >> --- /dev/null
> >> +++ b/drivers/block/nd/Makefile
> >> @@ -0,0 +1,6 @@
> >> +obj-$(CONFIG_LIBND) += libnd.o
> >> +obj-$(CONFIG_ND_ACPI) += nd_acpi.o
> >> +
> >> +nd_acpi-y := acpi.o
> >> +
> >> +libnd-y := core.o
> >
> > OK, so it looks like no modules, just built-in code, right?
> >
>
> Um, no, both CONFIG_ND_ACPI and CONFIG_LIBND can be =m.
OK
[cut]
> >> +static int nd_acpi_remove(struct acpi_device *adev)
> >> +{
> >> + struct acpi_nfit_desc *acpi_desc = dev_get_drvdata(&adev->dev);
> >> +
> >> + nd_bus_unregister(acpi_desc->nd_bus);
> >> + return 0;
> >> +}
> >> +
> >> +static void nd_acpi_notify(struct acpi_device *adev, u32 event)
> >> +{
> >> + /* TODO: handle ACPI_NOTIFY_BUS_CHECK notification */
> >> + dev_dbg(&adev->dev, "%s: event: %d\n", __func__, event);
> >> +}
> >> +
> >> +static const struct acpi_device_id nd_acpi_ids[] = {
> >> + { "ACPI0012", 0 },
> >> + { "", 0 },
> >> +};
> >> +MODULE_DEVICE_TABLE(acpi, nd_acpi_ids);
> >> +
> >> +static struct acpi_driver nd_acpi_driver = {
> >> + .name = KBUILD_MODNAME,
> >> + .ids = nd_acpi_ids,
> >> + .flags = ACPI_DRIVER_ALL_NOTIFY_EVENTS,
> >> + .ops = {
> >> + .add = nd_acpi_add,
> >> + .remove = nd_acpi_remove,
> >> + .notify = nd_acpi_notify
> >> + },
> >> +};
> >
> > Since this is going to be non-modular built-in code, please use an ACPI
> > scan handler instead of using a driver here. acpi_memhotplug.c does that,
> > you can use it as an example, but I guess you don't need to enable hotplug
> > for it to start with.
>
>
> No, you misunderstood, this will certainly be modular and loaded on-demand.
OK
So please drop the .notify thing at least for now. It most likely doesn't do
what you need anyway.
> >
> >> +
> >> +static __init int nd_acpi_init(void)
> >> +{
> >> + BUILD_BUG_ON(sizeof(struct acpi_nfit) != 40);
> >> + BUILD_BUG_ON(sizeof(struct acpi_nfit_spa) != 56);
> >> + BUILD_BUG_ON(sizeof(struct acpi_nfit_memdev) != 48);
> >> + BUILD_BUG_ON(sizeof(struct acpi_nfit_idt) != 16);
> >> + BUILD_BUG_ON(sizeof(struct acpi_nfit_smbios) != 8);
> >> + BUILD_BUG_ON(sizeof(struct acpi_nfit_dcr) != 80);
> >> + BUILD_BUG_ON(sizeof(struct acpi_nfit_bdw) != 40);
> >> +
> >> + return acpi_bus_register_driver(&nd_acpi_driver);
> >> +}
> >> +
> >> +static __exit void nd_acpi_exit(void)
> >> +{
> >> + acpi_bus_unregister_driver(&nd_acpi_driver);
> >> +}
> >> +
> >> +module_init(nd_acpi_init);
> >> +module_exit(nd_acpi_exit);
> >> +MODULE_LICENSE("GPL v2");
> >> +MODULE_AUTHOR("Intel Corporation");
> >> diff --git a/drivers/block/nd/acpi_nfit.h b/drivers/block/nd/acpi_nfit.h
> >> new file mode 100644
> >> index 000000000000..e0b0f12736bf
> >> --- /dev/null
> >> +++ b/drivers/block/nd/acpi_nfit.h
> >
> > I'm assuming that the below is coordinated with Bob and David and will be
> > changed to use ACPICA-provided definitions going forward.
> >
> > Is that correct?
>
> Yes, as soon as definitions those are available we will drop this
> header and rebase on the ACPICA implementation.
>
> [..]
OK
--
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.
On Thu, Apr 30, 2015 at 6:21 PM, Rafael J. Wysocki <[email protected]> wrote:
> On Thursday, April 30, 2015 05:39:06 PM Dan Williams wrote:
>> On Thu, Apr 30, 2015 at 4:23 PM, Rafael J. Wysocki <[email protected]> wrote:
[..]
>> >> +if ND_DEVICES
>> >> +
>> >> +config LIBND
>> >> + tristate "LIBND: libnd device driver support"
>> >> + help
>> >> + Platform agnostic device model for a libnd bus. Publishes
>> >> + resources for a PMEM (persistent-memory) driver and/or BLK
>> >> + (sliding mmio window(s)) driver to attach. Exposes a device
>> >> + topology under a "ndX" bus device, a "/dev/ndctlX" bus-ioctl
>> >> + message passing interface, and a "/dev/nmemX" dimm-ioctl
>> >> + message interface for each memory device registered on the
>> >> + bus. instance. A userspace library "ndctl" provides an API
>> >> + to enumerate/manage this subsystem.
>> >> +
>> >> +config ND_ACPI
>> >> + tristate "ACPI: NFIT to libnd bus support"
>> >> + select LIBND
>> >> + depends on ACPI
>> >> + help
>> >> + Infrastructure to probe ACPI 6 compliant platforms for
>> >> + NVDIMMs (NFIT) and register a libnd device tree. In
>> >> + addition to storage devices this also enables libnd craft
>> >> + ACPI._DSM messages for platform/dimm configuration.
>> >
>> > I'm wondering if the two CONFIG options above really need to be user-selectable?
>> >
>> > For example, what reason people (who've already selected ND_DEVICES) may have
>> > for not selecting ND_ACPI if ACPI is set?
>>
>>
>> Later on in the series we introduce ND_E820 which supports creating a
>> libnd-bus from e820-type-12 memory ranges on pre-NFIT systems. I'm
>> also considering a configfs defined libnd-bus because e820 types are
>> not nearly enough information to safely define nvdimm resources
>> outside of NFIT.
>
> I hope these are not mutually exclusive with ND_ACPI? Otherwise distros
> will have problems with supporting them in one kernel.
You can have ND_E820 support and ND_ACPI support in the same system.
Likely an NFIT enabled system will never have e820-type-12 ranges, but
if a user messes up and uses the new memmap=ss!nn command line to
overlap NFIT-defined memory then the request_mem_region() calls in the
driver will collide. First to load wins in that scenario.
> If ND_E820 and ND_ACPI aren't mutually exclusive, I still don't see a good
> enough reason for asking users about ND_ACPI. Why would I ever say "No"
> here if I said "Yes" or "Module" to ND_DEVICES?
I agree that if the user selects ND_DEVICES then ND_ACPI should
probably default on, but otherwise turning it off is a useful option.
If you know your system is pre-ACPI-6 then why bother including
support?
>> >> +
>> >> +endif
>> >> diff --git a/drivers/block/nd/Makefile b/drivers/block/nd/Makefile
>> >> new file mode 100644
>> >> index 000000000000..944b5947c0cb
>> >> --- /dev/null
>> >> +++ b/drivers/block/nd/Makefile
>> >> @@ -0,0 +1,6 @@
>> >> +obj-$(CONFIG_LIBND) += libnd.o
>> >> +obj-$(CONFIG_ND_ACPI) += nd_acpi.o
>> >> +
>> >> +nd_acpi-y := acpi.o
>> >> +
>> >> +libnd-y := core.o
>> >
>> > OK, so it looks like no modules, just built-in code, right?
>> >
>>
>> Um, no, both CONFIG_ND_ACPI and CONFIG_LIBND can be =m.
>
> OK
>
> [cut]
>
>> >> +static int nd_acpi_remove(struct acpi_device *adev)
>> >> +{
>> >> + struct acpi_nfit_desc *acpi_desc = dev_get_drvdata(&adev->dev);
>> >> +
>> >> + nd_bus_unregister(acpi_desc->nd_bus);
>> >> + return 0;
>> >> +}
>> >> +
>> >> +static void nd_acpi_notify(struct acpi_device *adev, u32 event)
>> >> +{
>> >> + /* TODO: handle ACPI_NOTIFY_BUS_CHECK notification */
>> >> + dev_dbg(&adev->dev, "%s: event: %d\n", __func__, event);
>> >> +}
>> >> +
>> >> +static const struct acpi_device_id nd_acpi_ids[] = {
>> >> + { "ACPI0012", 0 },
>> >> + { "", 0 },
>> >> +};
>> >> +MODULE_DEVICE_TABLE(acpi, nd_acpi_ids);
>> >> +
>> >> +static struct acpi_driver nd_acpi_driver = {
>> >> + .name = KBUILD_MODNAME,
>> >> + .ids = nd_acpi_ids,
>> >> + .flags = ACPI_DRIVER_ALL_NOTIFY_EVENTS,
>> >> + .ops = {
>> >> + .add = nd_acpi_add,
>> >> + .remove = nd_acpi_remove,
>> >> + .notify = nd_acpi_notify
>> >> + },
>> >> +};
>> >
>> > Since this is going to be non-modular built-in code, please use an ACPI
>> > scan handler instead of using a driver here. acpi_memhotplug.c does that,
>> > you can use it as an example, but I guess you don't need to enable hotplug
>> > for it to start with.
>>
>>
>> No, you misunderstood, this will certainly be modular and loaded on-demand.
>
> OK
>
> So please drop the .notify thing at least for now. It most likely doesn't do
> what you need anyway.
The .notify handler will eventually be filled in to handle hot-add of
NFIT structures, but yes I'll drop it for now.
On Tue, 2015-04-28 at 14:24 -0400, Dan Williams wrote:
> Register the memory devices described in the nfit as libnd 'dimm'
> devices on an nd bus. The kernel assigned device id for dimms is
> dynamic. If userspace needs a more static identifier it should consult
> a provider-specific attribute. In the case where NFIT is the provider,
> the 'nmemX/nfit/handle' or 'nmemX/nfit/serial' attributes may be used
> for this purpose.
:
> +
> +static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
> +{
> + struct nfit_mem *nfit_mem;
> +
> + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) {
> + struct nd_dimm *nd_dimm;
> + unsigned long flags = 0;
> + u32 nfit_handle;
> +
> + nfit_handle = __to_nfit_memdev(nfit_mem)->nfit_handle;
> + nd_dimm = nd_acpi_dimm_by_handle(acpi_desc, nfit_handle);
> + if (nd_dimm) {
> + /*
> + * If for some reason we find multiple DCRs the
> + * first one wins
> + */
> + dev_err(acpi_desc->dev, "duplicate DCR detected: %s\n",
> + nd_dimm_name(nd_dimm));
> + continue;
> + }
> +
> + if (nfit_mem->bdw && nfit_mem->memdev_pmem)
> + flags |= NDD_ALIASING;
Does this check work for a NVDIMM card which has multiple pmem regions
with label info, but does not have any bdw region configured?
The code assumes that namespace_pmem (NDD_ALIASING) and namespace_blk
have label info. There may be an NVDIMM card with a single blk region
without label info.
Instead of using the namespace types to assume the label info, how about
adding a flag to indicate the presence of the label info? This avoids
the separation of namespace_io and namespace_pmem for the same pmem
driver.
Thanks,
-Toshi
On Fri, 2015-05-01 at 11:22 -0700, Dan Williams wrote:
> On Fri, May 1, 2015 at 10:48 AM, Toshi Kani <[email protected]> wrote:
> > On Tue, 2015-04-28 at 14:24 -0400, Dan Williams wrote:
> >> Register the memory devices described in the nfit as libnd 'dimm'
> >> devices on an nd bus. The kernel assigned device id for dimms is
> >> dynamic. If userspace needs a more static identifier it should consult
> >> a provider-specific attribute. In the case where NFIT is the provider,
> >> the 'nmemX/nfit/handle' or 'nmemX/nfit/serial' attributes may be used
> >> for this purpose.
> > :
> >> +
> >> +static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
> >> +{
> >> + struct nfit_mem *nfit_mem;
> >> +
> >> + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) {
> >> + struct nd_dimm *nd_dimm;
> >> + unsigned long flags = 0;
> >> + u32 nfit_handle;
> >> +
> >> + nfit_handle = __to_nfit_memdev(nfit_mem)->nfit_handle;
> >> + nd_dimm = nd_acpi_dimm_by_handle(acpi_desc, nfit_handle);
> >> + if (nd_dimm) {
> >> + /*
> >> + * If for some reason we find multiple DCRs the
> >> + * first one wins
> >> + */
> >> + dev_err(acpi_desc->dev, "duplicate DCR detected: %s\n",
> >> + nd_dimm_name(nd_dimm));
> >> + continue;
> >> + }
> >> +
> >> + if (nfit_mem->bdw && nfit_mem->memdev_pmem)
> >> + flags |= NDD_ALIASING;
> >
> > Does this check work for a NVDIMM card which has multiple pmem regions
> > with label info, but does not have any bdw region configured?
>
> If you have multiple pmem regions then you don't have aliasing and
> don't need a label. You'll get an nd_namespace_io per region.
>
> > The code assumes that namespace_pmem (NDD_ALIASING) and namespace_blk
> > have label info. There may be an NVDIMM card with a single blk region
> > without label info.
>
> I'd really like to suggest that labels are only for resolving aliasing
> and that if you have a BLK-only NVDIMM you'll get an automatic
> namespace created the same as a PMEM-only. Partitioning is always
> there to provide sub-divisions of a namespace. The only reason to
> support multiple BLK-namespaces per-region is to give each a different
> sector size. I may eventually need to relent on this position, but
> I'd really like to understand the use case for requiring labels when
> aliasing is not present as it seems like a waste to me.
By looking at the callers of is_namespace_pmem() and is_namespace_blk(),
such as nd_namespace_label_update(), I am concerned that the namespace
types are also used for indicating the presence a label. Is it OK for
nd_namespace_label_update() to do nothing when there is no aliasing?
> > Instead of using the namespace types to assume the label info, how about
> > adding a flag to indicate the presence of the label info? This avoids
> > the separation of namespace_io and namespace_pmem for the same pmem
> > driver.
>
> To what benefit?
Why do they need to be separated? Having alias or not should not make
the pmem namespace different.
Thanks,
-Toshi
On Fri, May 1, 2015 at 10:48 AM, Toshi Kani <[email protected]> wrote:
> On Tue, 2015-04-28 at 14:24 -0400, Dan Williams wrote:
>> Register the memory devices described in the nfit as libnd 'dimm'
>> devices on an nd bus. The kernel assigned device id for dimms is
>> dynamic. If userspace needs a more static identifier it should consult
>> a provider-specific attribute. In the case where NFIT is the provider,
>> the 'nmemX/nfit/handle' or 'nmemX/nfit/serial' attributes may be used
>> for this purpose.
> :
>> +
>> +static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
>> +{
>> + struct nfit_mem *nfit_mem;
>> +
>> + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) {
>> + struct nd_dimm *nd_dimm;
>> + unsigned long flags = 0;
>> + u32 nfit_handle;
>> +
>> + nfit_handle = __to_nfit_memdev(nfit_mem)->nfit_handle;
>> + nd_dimm = nd_acpi_dimm_by_handle(acpi_desc, nfit_handle);
>> + if (nd_dimm) {
>> + /*
>> + * If for some reason we find multiple DCRs the
>> + * first one wins
>> + */
>> + dev_err(acpi_desc->dev, "duplicate DCR detected: %s\n",
>> + nd_dimm_name(nd_dimm));
>> + continue;
>> + }
>> +
>> + if (nfit_mem->bdw && nfit_mem->memdev_pmem)
>> + flags |= NDD_ALIASING;
>
> Does this check work for a NVDIMM card which has multiple pmem regions
> with label info, but does not have any bdw region configured?
If you have multiple pmem regions then you don't have aliasing and
don't need a label. You'll get an nd_namespace_io per region.
> The code assumes that namespace_pmem (NDD_ALIASING) and namespace_blk
> have label info. There may be an NVDIMM card with a single blk region
> without label info.
I'd really like to suggest that labels are only for resolving aliasing
and that if you have a BLK-only NVDIMM you'll get an automatic
namespace created the same as a PMEM-only. Partitioning is always
there to provide sub-divisions of a namespace. The only reason to
support multiple BLK-namespaces per-region is to give each a different
sector size. I may eventually need to relent on this position, but
I'd really like to understand the use case for requiring labels when
aliasing is not present as it seems like a waste to me.
> Instead of using the namespace types to assume the label info, how about
> adding a flag to indicate the presence of the label info? This avoids
> the separation of namespace_io and namespace_pmem for the same pmem
> driver.
To what benefit?
On Fri, May 1, 2015 at 11:19 AM, Toshi Kani <[email protected]> wrote:
> On Fri, 2015-05-01 at 11:22 -0700, Dan Williams wrote:
>> On Fri, May 1, 2015 at 10:48 AM, Toshi Kani <[email protected]> wrote:
>> > On Tue, 2015-04-28 at 14:24 -0400, Dan Williams wrote:
>> >> Register the memory devices described in the nfit as libnd 'dimm'
>> >> devices on an nd bus. The kernel assigned device id for dimms is
>> >> dynamic. If userspace needs a more static identifier it should consult
>> >> a provider-specific attribute. In the case where NFIT is the provider,
>> >> the 'nmemX/nfit/handle' or 'nmemX/nfit/serial' attributes may be used
>> >> for this purpose.
>> > :
>> >> +
>> >> +static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
>> >> +{
>> >> + struct nfit_mem *nfit_mem;
>> >> +
>> >> + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) {
>> >> + struct nd_dimm *nd_dimm;
>> >> + unsigned long flags = 0;
>> >> + u32 nfit_handle;
>> >> +
>> >> + nfit_handle = __to_nfit_memdev(nfit_mem)->nfit_handle;
>> >> + nd_dimm = nd_acpi_dimm_by_handle(acpi_desc, nfit_handle);
>> >> + if (nd_dimm) {
>> >> + /*
>> >> + * If for some reason we find multiple DCRs the
>> >> + * first one wins
>> >> + */
>> >> + dev_err(acpi_desc->dev, "duplicate DCR detected: %s\n",
>> >> + nd_dimm_name(nd_dimm));
>> >> + continue;
>> >> + }
>> >> +
>> >> + if (nfit_mem->bdw && nfit_mem->memdev_pmem)
>> >> + flags |= NDD_ALIASING;
>> >
>> > Does this check work for a NVDIMM card which has multiple pmem regions
>> > with label info, but does not have any bdw region configured?
>>
>> If you have multiple pmem regions then you don't have aliasing and
>> don't need a label. You'll get an nd_namespace_io per region.
>>
>> > The code assumes that namespace_pmem (NDD_ALIASING) and namespace_blk
>> > have label info. There may be an NVDIMM card with a single blk region
>> > without label info.
>>
>> I'd really like to suggest that labels are only for resolving aliasing
>> and that if you have a BLK-only NVDIMM you'll get an automatic
>> namespace created the same as a PMEM-only. Partitioning is always
>> there to provide sub-divisions of a namespace. The only reason to
>> support multiple BLK-namespaces per-region is to give each a different
>> sector size. I may eventually need to relent on this position, but
>> I'd really like to understand the use case for requiring labels when
>> aliasing is not present as it seems like a waste to me.
>
> By looking at the callers of is_namespace_pmem() and is_namespace_blk(),
> such as nd_namespace_label_update(), I am concerned that the namespace
> types are also used for indicating the presence a label. Is it OK for
> nd_namespace_label_update() to do nothing when there is no aliasing?
>
>> > Instead of using the namespace types to assume the label info, how about
>> > adding a flag to indicate the presence of the label info? This avoids
>> > the separation of namespace_io and namespace_pmem for the same pmem
>> > driver.
>>
>> To what benefit?
>
> Why do they need to be separated? Having alias or not should not make
> the pmem namespace different.
The intent is to maximize the number of devices that can be
immediately attached to nd_pmem and nd_blk without user intervention.
nd_namespace_io is a pmem namespace where the boundaries are 100%
described by the NFIT / parent-region.
On Fri, 2015-05-01 at 11:43 -0700, Dan Williams wrote:
> On Fri, May 1, 2015 at 11:19 AM, Toshi Kani <[email protected]> wrote:
> > On Fri, 2015-05-01 at 11:22 -0700, Dan Williams wrote:
> >> On Fri, May 1, 2015 at 10:48 AM, Toshi Kani <[email protected]> wrote:
> >> > On Tue, 2015-04-28 at 14:24 -0400, Dan Williams wrote:
> >> >> Register the memory devices described in the nfit as libnd 'dimm'
> >> >> devices on an nd bus. The kernel assigned device id for dimms is
> >> >> dynamic. If userspace needs a more static identifier it should consult
> >> >> a provider-specific attribute. In the case where NFIT is the provider,
> >> >> the 'nmemX/nfit/handle' or 'nmemX/nfit/serial' attributes may be used
> >> >> for this purpose.
> >> > :
> >> >> +
> >> >> +static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
> >> >> +{
> >> >> + struct nfit_mem *nfit_mem;
> >> >> +
> >> >> + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) {
> >> >> + struct nd_dimm *nd_dimm;
> >> >> + unsigned long flags = 0;
> >> >> + u32 nfit_handle;
> >> >> +
> >> >> + nfit_handle = __to_nfit_memdev(nfit_mem)->nfit_handle;
> >> >> + nd_dimm = nd_acpi_dimm_by_handle(acpi_desc, nfit_handle);
> >> >> + if (nd_dimm) {
> >> >> + /*
> >> >> + * If for some reason we find multiple DCRs the
> >> >> + * first one wins
> >> >> + */
> >> >> + dev_err(acpi_desc->dev, "duplicate DCR detected: %s\n",
> >> >> + nd_dimm_name(nd_dimm));
> >> >> + continue;
> >> >> + }
> >> >> +
> >> >> + if (nfit_mem->bdw && nfit_mem->memdev_pmem)
> >> >> + flags |= NDD_ALIASING;
> >> >
> >> > Does this check work for a NVDIMM card which has multiple pmem regions
> >> > with label info, but does not have any bdw region configured?
> >>
> >> If you have multiple pmem regions then you don't have aliasing and
> >> don't need a label. You'll get an nd_namespace_io per region.
> >>
> >> > The code assumes that namespace_pmem (NDD_ALIASING) and namespace_blk
> >> > have label info. There may be an NVDIMM card with a single blk region
> >> > without label info.
> >>
> >> I'd really like to suggest that labels are only for resolving aliasing
> >> and that if you have a BLK-only NVDIMM you'll get an automatic
> >> namespace created the same as a PMEM-only. Partitioning is always
> >> there to provide sub-divisions of a namespace. The only reason to
> >> support multiple BLK-namespaces per-region is to give each a different
> >> sector size. I may eventually need to relent on this position, but
> >> I'd really like to understand the use case for requiring labels when
> >> aliasing is not present as it seems like a waste to me.
> >
> > By looking at the callers of is_namespace_pmem() and is_namespace_blk(),
> > such as nd_namespace_label_update(), I am concerned that the namespace
> > types are also used for indicating the presence a label. Is it OK for
> > nd_namespace_label_update() to do nothing when there is no aliasing?
Did you forget to answer this question? I am not asking to have a
label. I am asking if the namespace types can handle it correctly.
Restating the nd_namespace_label_update() example:
- namespace_io case: Skip, but a label may still exist. Correct?
- namespace_blk case: Proceed, but blk does not require a label.
> >> > Instead of using the namespace types to assume the label info, how about
> >> > adding a flag to indicate the presence of the label info? This avoids
> >> > the separation of namespace_io and namespace_pmem for the same pmem
> >> > driver.
> >>
> >> To what benefit?
> >
> > Why do they need to be separated? Having alias or not should not make
> > the pmem namespace different.
>
> The intent is to maximize the number of devices that can be
> immediately attached to nd_pmem and nd_blk without user intervention.
I agree with your intention. Again, I am not asking to have a label.
> nd_namespace_io is a pmem namespace where the boundaries are 100%
> described by the NFIT / parent-region.
Thanks,
-Toshi
On Fri, May 1, 2015 at 12:15 PM, Toshi Kani <[email protected]> wrote:
> On Fri, 2015-05-01 at 11:43 -0700, Dan Williams wrote:
>> On Fri, May 1, 2015 at 11:19 AM, Toshi Kani <[email protected]> wrote:
>> > On Fri, 2015-05-01 at 11:22 -0700, Dan Williams wrote:
>> >> On Fri, May 1, 2015 at 10:48 AM, Toshi Kani <[email protected]> wrote:
>> >> > On Tue, 2015-04-28 at 14:24 -0400, Dan Williams wrote:
>> >> >> Register the memory devices described in the nfit as libnd 'dimm'
>> >> >> devices on an nd bus. The kernel assigned device id for dimms is
>> >> >> dynamic. If userspace needs a more static identifier it should consult
>> >> >> a provider-specific attribute. In the case where NFIT is the provider,
>> >> >> the 'nmemX/nfit/handle' or 'nmemX/nfit/serial' attributes may be used
>> >> >> for this purpose.
>> >> > :
>> >> >> +
>> >> >> +static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
>> >> >> +{
>> >> >> + struct nfit_mem *nfit_mem;
>> >> >> +
>> >> >> + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) {
>> >> >> + struct nd_dimm *nd_dimm;
>> >> >> + unsigned long flags = 0;
>> >> >> + u32 nfit_handle;
>> >> >> +
>> >> >> + nfit_handle = __to_nfit_memdev(nfit_mem)->nfit_handle;
>> >> >> + nd_dimm = nd_acpi_dimm_by_handle(acpi_desc, nfit_handle);
>> >> >> + if (nd_dimm) {
>> >> >> + /*
>> >> >> + * If for some reason we find multiple DCRs the
>> >> >> + * first one wins
>> >> >> + */
>> >> >> + dev_err(acpi_desc->dev, "duplicate DCR detected: %s\n",
>> >> >> + nd_dimm_name(nd_dimm));
>> >> >> + continue;
>> >> >> + }
>> >> >> +
>> >> >> + if (nfit_mem->bdw && nfit_mem->memdev_pmem)
>> >> >> + flags |= NDD_ALIASING;
>> >> >
>> >> > Does this check work for a NVDIMM card which has multiple pmem regions
>> >> > with label info, but does not have any bdw region configured?
>> >>
>> >> If you have multiple pmem regions then you don't have aliasing and
>> >> don't need a label. You'll get an nd_namespace_io per region.
>> >>
>> >> > The code assumes that namespace_pmem (NDD_ALIASING) and namespace_blk
>> >> > have label info. There may be an NVDIMM card with a single blk region
>> >> > without label info.
>> >>
>> >> I'd really like to suggest that labels are only for resolving aliasing
>> >> and that if you have a BLK-only NVDIMM you'll get an automatic
>> >> namespace created the same as a PMEM-only. Partitioning is always
>> >> there to provide sub-divisions of a namespace. The only reason to
>> >> support multiple BLK-namespaces per-region is to give each a different
>> >> sector size. I may eventually need to relent on this position, but
>> >> I'd really like to understand the use case for requiring labels when
>> >> aliasing is not present as it seems like a waste to me.
>> >
>> > By looking at the callers of is_namespace_pmem() and is_namespace_blk(),
>> > such as nd_namespace_label_update(), I am concerned that the namespace
>> > types are also used for indicating the presence a label. Is it OK for
>> > nd_namespace_label_update() to do nothing when there is no aliasing?
>
> Did you forget to answer this question? I am not asking to have a
> label. I am asking if the namespace types can handle it correctly.
> Restating the nd_namespace_label_update() example:
> - namespace_io case: Skip, but a label may still exist. Correct?
> - namespace_blk case: Proceed, but blk does not require a label.
Ah, ok. This is handled by nd_namespace_attr_visible() only labelled
namespaces have writable sysfs attributes. This would need to be
extended for a label-less BLK namespace type.
On Fri, 2015-05-01 at 12:38 -0700, Dan Williams wrote:
> On Fri, May 1, 2015 at 12:15 PM, Toshi Kani <[email protected]> wrote:
> > On Fri, 2015-05-01 at 11:43 -0700, Dan Williams wrote:
> >> On Fri, May 1, 2015 at 11:19 AM, Toshi Kani <[email protected]> wrote:
> >> > On Fri, 2015-05-01 at 11:22 -0700, Dan Williams wrote:
> >> >> On Fri, May 1, 2015 at 10:48 AM, Toshi Kani <[email protected]> wrote:
> >> >> > On Tue, 2015-04-28 at 14:24 -0400, Dan Williams wrote:
> >> >> >> Register the memory devices described in the nfit as libnd 'dimm'
> >> >> >> devices on an nd bus. The kernel assigned device id for dimms is
> >> >> >> dynamic. If userspace needs a more static identifier it should consult
> >> >> >> a provider-specific attribute. In the case where NFIT is the provider,
> >> >> >> the 'nmemX/nfit/handle' or 'nmemX/nfit/serial' attributes may be used
> >> >> >> for this purpose.
> >> >> > :
> >> >> >> +
> >> >> >> +static int nd_acpi_register_dimms(struct acpi_nfit_desc *acpi_desc)
> >> >> >> +{
> >> >> >> + struct nfit_mem *nfit_mem;
> >> >> >> +
> >> >> >> + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) {
> >> >> >> + struct nd_dimm *nd_dimm;
> >> >> >> + unsigned long flags = 0;
> >> >> >> + u32 nfit_handle;
> >> >> >> +
> >> >> >> + nfit_handle = __to_nfit_memdev(nfit_mem)->nfit_handle;
> >> >> >> + nd_dimm = nd_acpi_dimm_by_handle(acpi_desc, nfit_handle);
> >> >> >> + if (nd_dimm) {
> >> >> >> + /*
> >> >> >> + * If for some reason we find multiple DCRs the
> >> >> >> + * first one wins
> >> >> >> + */
> >> >> >> + dev_err(acpi_desc->dev, "duplicate DCR detected: %s\n",
> >> >> >> + nd_dimm_name(nd_dimm));
> >> >> >> + continue;
> >> >> >> + }
> >> >> >> +
> >> >> >> + if (nfit_mem->bdw && nfit_mem->memdev_pmem)
> >> >> >> + flags |= NDD_ALIASING;
> >> >> >
> >> >> > Does this check work for a NVDIMM card which has multiple pmem regions
> >> >> > with label info, but does not have any bdw region configured?
> >> >>
> >> >> If you have multiple pmem regions then you don't have aliasing and
> >> >> don't need a label. You'll get an nd_namespace_io per region.
> >> >>
> >> >> > The code assumes that namespace_pmem (NDD_ALIASING) and namespace_blk
> >> >> > have label info. There may be an NVDIMM card with a single blk region
> >> >> > without label info.
> >> >>
> >> >> I'd really like to suggest that labels are only for resolving aliasing
> >> >> and that if you have a BLK-only NVDIMM you'll get an automatic
> >> >> namespace created the same as a PMEM-only. Partitioning is always
> >> >> there to provide sub-divisions of a namespace. The only reason to
> >> >> support multiple BLK-namespaces per-region is to give each a different
> >> >> sector size. I may eventually need to relent on this position, but
> >> >> I'd really like to understand the use case for requiring labels when
> >> >> aliasing is not present as it seems like a waste to me.
> >> >
> >> > By looking at the callers of is_namespace_pmem() and is_namespace_blk(),
> >> > such as nd_namespace_label_update(), I am concerned that the namespace
> >> > types are also used for indicating the presence a label. Is it OK for
> >> > nd_namespace_label_update() to do nothing when there is no aliasing?
> >
> > Did you forget to answer this question? I am not asking to have a
> > label. I am asking if the namespace types can handle it correctly.
> > Restating the nd_namespace_label_update() example:
> > - namespace_io case: Skip, but a label may still exist. Correct?
> > - namespace_blk case: Proceed, but blk does not require a label.
>
> Ah, ok. This is handled by nd_namespace_attr_visible() only labelled
> namespaces have writable sysfs attributes. This would need to be
> extended for a label-less BLK namespace type.
I prefer not to duplicate each namespace type with and without a label,
but I am OK as long as the presence of labels is handled properly.
Thanks,
-Toshi
On Tue, 2015-04-28 at 14:24 -0400, Dan Williams wrote:
:
> +
> +static int nd_acpi_register_region(struct acpi_nfit_desc *acpi_desc,
> + struct nfit_spa *nfit_spa)
> +{
> + static struct nd_mapping nd_mappings[ND_MAX_MAPPINGS];
> + struct acpi_nfit_spa *spa = nfit_spa->spa;
> + struct nfit_memdev *nfit_memdev;
> + struct nd_region_desc ndr_desc;
> + int spa_type, count = 0;
> + struct resource res;
> + u16 spa_index;
> +
> + spa_type = nfit_spa_type(spa);
> + spa_index = spa->spa_index;
> + if (spa_index == 0) {
> + dev_dbg(acpi_desc->dev, "%s: detected invalid spa index\n",
> + __func__);
> + return 0;
> + }
> +
> + memset(&res, 0, sizeof(res));
> + memset(&nd_mappings, 0, sizeof(nd_mappings));
> + memset(&ndr_desc, 0, sizeof(ndr_desc));
> + res.start = spa->spa_base;
> + res.end = res.start + spa->spa_length - 1;
> + ndr_desc.res = &res;
> + ndr_desc.provider_data = nfit_spa;
> + ndr_desc.attr_groups = nd_acpi_region_attribute_groups;
> + list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) {
> + struct acpi_nfit_memdev *memdev = nfit_memdev->memdev;
> + struct nd_mapping *nd_mapping;
> + struct nd_dimm *nd_dimm;
> +
> + if (memdev->spa_index != spa_index)
> + continue;
The libnd does not support memdev->flags, which contains "Memory Device
State Flags" defined in Table 5-129 of ACPI 6.0. In case of major
errors, we should only allow a failed NVDIMM be accessed with read-only
for possible data recovery (or not allow any access when the data is
completely lost), and should not let users operate normally over the
corrupted data until the error is dealt properly.
Can you set memdev->flags to nd_region(_desc) so that the pmem driver
can check the status in nd_pmem_probe()? nd_pmem_probe() can then set
the disk read-only or fail probing, and log errors accordingly.
Thanks,
-Toshi
On Mon, May 4, 2015 at 4:58 PM, Rafael J. Wysocki <[email protected]> wrote:
> On Friday, May 01, 2015 09:23:38 AM Dan Williams wrote:
>> On Thu, Apr 30, 2015 at 6:21 PM, Rafael J. Wysocki <[email protected]> wrote:
>> > On Thursday, April 30, 2015 05:39:06 PM Dan Williams wrote:
>> >> On Thu, Apr 30, 2015 at 4:23 PM, Rafael J. Wysocki <[email protected]> wrote:
[..]
>> > If ND_E820 and ND_ACPI aren't mutually exclusive, I still don't see a good
>> > enough reason for asking users about ND_ACPI. Why would I ever say "No"
>> > here if I said "Yes" or "Module" to ND_DEVICES?
>>
>> I agree that if the user selects ND_DEVICES then ND_ACPI should
>> probably default on, but otherwise turning it off is a useful option.
>> If you know your system is pre-ACPI-6 then why bother including
>> support?
>
> If you're a distro, you don't care. You have to support it regardless.
>
> You might care if you're an end user building a kernel for yourself and just
> for this particular specific machine. Honestly, how many *server* users do
> that?
>
> And fewer user-selectable options means fewer combination of options to test
> during development/validation.
>
> Also unrelated, but applies to this patch.
>
> Since your new driver will handle device ID ACPI0012 which is defined by the
> spec proper, it should go into drivers/acpi/, because there's where such things
> go as a rule.
Ok, I think the move to drivers/acpi/ will kill two birds with one
stone as selecting ACPI_NFIT from there will select the libnd support
without prompting.
On Friday, May 01, 2015 09:23:38 AM Dan Williams wrote:
> On Thu, Apr 30, 2015 at 6:21 PM, Rafael J. Wysocki <[email protected]> wrote:
> > On Thursday, April 30, 2015 05:39:06 PM Dan Williams wrote:
> >> On Thu, Apr 30, 2015 at 4:23 PM, Rafael J. Wysocki <[email protected]> wrote:
> [..]
> >> >> +if ND_DEVICES
> >> >> +
> >> >> +config LIBND
> >> >> + tristate "LIBND: libnd device driver support"
> >> >> + help
> >> >> + Platform agnostic device model for a libnd bus. Publishes
> >> >> + resources for a PMEM (persistent-memory) driver and/or BLK
> >> >> + (sliding mmio window(s)) driver to attach. Exposes a device
> >> >> + topology under a "ndX" bus device, a "/dev/ndctlX" bus-ioctl
> >> >> + message passing interface, and a "/dev/nmemX" dimm-ioctl
> >> >> + message interface for each memory device registered on the
> >> >> + bus. instance. A userspace library "ndctl" provides an API
> >> >> + to enumerate/manage this subsystem.
> >> >> +
> >> >> +config ND_ACPI
> >> >> + tristate "ACPI: NFIT to libnd bus support"
> >> >> + select LIBND
> >> >> + depends on ACPI
> >> >> + help
> >> >> + Infrastructure to probe ACPI 6 compliant platforms for
> >> >> + NVDIMMs (NFIT) and register a libnd device tree. In
> >> >> + addition to storage devices this also enables libnd craft
> >> >> + ACPI._DSM messages for platform/dimm configuration.
> >> >
> >> > I'm wondering if the two CONFIG options above really need to be user-selectable?
> >> >
> >> > For example, what reason people (who've already selected ND_DEVICES) may have
> >> > for not selecting ND_ACPI if ACPI is set?
> >>
> >>
> >> Later on in the series we introduce ND_E820 which supports creating a
> >> libnd-bus from e820-type-12 memory ranges on pre-NFIT systems. I'm
> >> also considering a configfs defined libnd-bus because e820 types are
> >> not nearly enough information to safely define nvdimm resources
> >> outside of NFIT.
> >
> > I hope these are not mutually exclusive with ND_ACPI? Otherwise distros
> > will have problems with supporting them in one kernel.
>
> You can have ND_E820 support and ND_ACPI support in the same system.
> Likely an NFIT enabled system will never have e820-type-12 ranges, but
> if a user messes up and uses the new memmap=ss!nn command line to
> overlap NFIT-defined memory then the request_mem_region() calls in the
> driver will collide. First to load wins in that scenario.
>
> > If ND_E820 and ND_ACPI aren't mutually exclusive, I still don't see a good
> > enough reason for asking users about ND_ACPI. Why would I ever say "No"
> > here if I said "Yes" or "Module" to ND_DEVICES?
>
> I agree that if the user selects ND_DEVICES then ND_ACPI should
> probably default on, but otherwise turning it off is a useful option.
> If you know your system is pre-ACPI-6 then why bother including
> support?
If you're a distro, you don't care. You have to support it regardless.
You might care if you're an end user building a kernel for yourself and just
for this particular specific machine. Honestly, how many *server* users do
that?
And fewer user-selectable options means fewer combination of options to test
during development/validation.
Also unrelated, but applies to this patch.
Since your new driver will handle device ID ACPI0012 which is defined by the
spec proper, it should go into drivers/acpi/, because there's where such things
go as a rule.
--
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.
On Tuesday, April 28, 2015 06:22:05 PM Dan Williams wrote:
> On Tue, Apr 28, 2015 at 5:25 PM, Rafael J. Wysocki <[email protected]> wrote:
> > On Tuesday, April 28, 2015 02:24:12 PM Dan Williams wrote:
> >> Changes since v1 [1]: Incorporates feedback received prior to April 24.
> >>
[cut]
> >
> > I'm wondering what's wrong with CCing all of the series to linux-acpi?
> >
> > Is there anything in it that the people on that list should not see, by any
> > chance?
>
> linux-acpi may not care about the dimm-metadata labeling patches that
> are completely independent of ACPI, but might as well include
> linux-acpi on the whole series at this point.
I've gone through the ACPI-related patches in this series (other than [2/20]
that I've commented directly) and while I haven't found anything horrible in
them, I don't quite feel confident enough to ACK them.
What I'm really missing in this series is a design document describing all that
from a high-level perspective and making it clear where all of the pieces go
and what their respective roles are. Also reordering the series to introduce
the nd subsystem to start with and then its users might help here.
--
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.
On Tue, Apr 28, 2015 at 03:15:54PM -0700, Dan Williams wrote:
> > lsblk's blkdev_scsi_type_to_name() considers 4 to mean
> > SCSI_TYPE_WORM (write once read many ... used for certain optical
> > and tape drives).
>
> Why is lsblk assuming these are scsi devices? I'll need to go check that out.
It's a very common assumption unfortunately. I rember fixing it in
various in-house tools at customers and stumbled over it in targetcli
recently.
Please use a prefix for your type attribute to avoid this problem.
On Tue, 2015-05-05 at 02:06 +0200, Rafael J. Wysocki wrote:
> On Tuesday, April 28, 2015 06:22:05 PM Dan Williams wrote:
> > On Tue, Apr 28, 2015 at 5:25 PM, Rafael J. Wysocki <[email protected]> wrote:
> > > On Tuesday, April 28, 2015 02:24:12 PM Dan Williams wrote:
> > >> Changes since v1 [1]: Incorporates feedback received prior to April 24.
> > >>
>
> [cut]
>
> > >
> > > I'm wondering what's wrong with CCing all of the series to linux-acpi?
> > >
> > > Is there anything in it that the people on that list should not see, by any
> > > chance?
> >
> > linux-acpi may not care about the dimm-metadata labeling patches that
> > are completely independent of ACPI, but might as well include
> > linux-acpi on the whole series at this point.
>
> I've gone through the ACPI-related patches in this series (other than [2/20]
> that I've commented directly) and while I haven't found anything horrible in
> them, I don't quite feel confident enough to ACK them.
>
> What I'm really missing in this series is a design document describing all that
> from a high-level perspective and making it clear where all of the pieces go
> and what their respective roles are. Also reordering the series to introduce
> the nd subsystem to start with and then its users might help here.
Here you go, and also see the "Supporting Documents" section if you need
more details, or just ask. This is the reworked document after pushing
NFIT specifics out of the core implementation. The core apis are
nd_bus_register(), nd_dimm_create(), nd_pmem_region_create(), and
nd_blk_region_create().
---
LIBND: Non-volatile Devices
libnd - kernel / libndctl - userspace helper library
[email protected]
v10
Glossary
Overview
Supporting Documents
Git Trees
LIBND PMEM and BLK
Why BLK?
PMEM vs BLK
BLK-REGIONs, PMEM-REGIONs, Atomic Sectors, and DAX
Example NVDIMM Platform
LIBND Kernel Device Model and LIBNDCTL Userspace API
LIBNDCTL: Context
libndctl: instantiate a new library context example
LIBND/LIBNDCTL: Bus
libnd: control class device in /sys/class
libnd: bus
libndctl: bus enumeration example
LIBND/LIBNDCTL: DIMM (NMEM)
libnd: DIMM (NMEM)
libndctl: DIMM enumeration example
LIBND/LIBNDCTL: Region
libnd: region
libndctl: region enumeration example
Why Not Encode the Region Type into the Region Name?
How Do I Determine the Major Type of a Region?
LIBND/LIBNDCTL: Namespace
libnd: namespace
libndctl: namespace enumeration example
libndctl: namespace creation example
Why the Term "namespace"?
LIBND/LIBNDCTL: Block Translation Table "btt"
libnd: btt layout
libndctl: btt creation example
Summary LIBNDCTL Diagram
Glossary
--------
PMEM: A system physical address range where writes are persistent. A
block device composed of PMEM is capable of DAX. A PMEM address range
may span/interleave several DIMMs.
BLK: A set of one or more programmable memory mapped apertures provided
by a DIMM to access its media. This indirection precludes the
performance benefit of interleaving, but enables DIMM-bounded failure
modes .
DPA: DIMM Physical Address, is a DIMM-relative offset. With one DIMM in
the system there would be a 1:1 system-physical-address:DPA association.
Once more DIMMs are added an memory controller interleave must be
decoded to determine the DPA associated with a given
system-physical-address. BLK capacity always has a 1:1 relationship
with a single-dimm's DPA range.
DAX: File system extensions to bypass the page cache and block layer to
mmap persistent memory, from a PMEM block device, directly into a
process address space.
BTT: Block Translation Table: Persistent memory is byte addressable.
Existing software may have an expectation that the power-fail-atomicity
of writes is at least one sector, 512 bytes. The BTT is an indirection
table with atomic update semantics to front a PMEM/BLK block device
driver and present arbitrary atomic sector sizes.
LABEL: Metadata stored on a DIMM device that partitions and identifies
(persistently names) storage between PMEM and BLK. It also partitions
BLK storage to host BTTs with different parameters per BLK-partition.
Note that traditional partition tables, GPT/MBR, are layered on top of a
BLK or PMEM device.
Overview
--------
The libnd subsystem provides support for three types of NVDIMMs, PMEM,
BLK, and NVDIMM platforms that can simultaneously support PMEM and BLK
mode access capabilities on a given set of DIMMs. These three modes of
operation are described by the "NVDIMM Firmware Interface Table" (NFIT)
in ACPI 6. While the libnd implementation is generic and supports
pre-NFIT platforms, it was guided by the superset of capabilities need
to support this ACPI 6 definition for NVDIMM resources. The bulk of the
kernel implementation is in place to handle the case where DPA
accessible via PMEM is aliased with DPA accessible via BLK. When that
occurs a LABEL is needed to reserve DPA for exclusive access via one
mode a time.
Supporting Documents
ACPI 6: http://www.uefi.org/sites/default/files/resources/ACPI_6.0.pdf
NVDIMM Namespace: http://pmem.io/documents/NVDIMM_Namespace_Spec.pdf
DSM Interface Example: http://pmem.io/documents/NVDIMM_DSM_Interface_Example.pdf
Driver Writer's Guide: http://pmem.io/documents/NVDIMM_Driver_Writers_Guide.pdf
Git Trees
LIBND: https://git.kernel.org/cgit/linux/kernel/git/djbw/nvdimm.git/log/?h=nd
LIBNDCTL: https://github.com/pmem/ndctl.git
PMEM: https://github.com/01org/prd
LIBND PMEM and BLK
------------------
Prior to the arrival of the NFIT, non-volatile memory was described to a
system in various ad-hoc ways. Usually only the bare minimum was
provided, namely, a single system-physical-address range where writes
are expected to be durable after a system power loss. Now, the NFIT
specification standardizes not only the description of PMEM, but also
BLK and platform message-passing entry points for control and
configuration.
For each NVDIMM access method (PMEM, BLK), LIBND provides a block device driver:
1. PMEM (nd_pmem.ko): Drives a system-physical-address range. This
range is contiguous in system memory and may be interleaved (hardware
memory controller striped) across multiple DIMMs. When interleaved the
platform may optionally provide details of which DIMMs are participating
in the interleave.
Note, LIBND describes system-physical-address ranges that may alias with
BLK access ND_NAMESPACE_PMEM ranges and those without alias as
ND_NAMESPACE_IO ranges, to the nd_pmem driver there is no distinction.
The different device-types are an implementation detail that userspace
can exploit to implement policies like "only interface with address
ranges from certain DIMMs". It is worth noting that when aliasing is
present and a DIMM lacks a label, then no block device can be created by
default as userspace needs to do at least one allocation of DPA to the
PMEM range. In contrast ND_NAMESPACE_IO ranges, once registered, can be
immediately attached to nd_pmem.
2. BLK (nd_blk.ko): This driver performs I/O using a set of platform
defined apertures. A set of apertures will all access just one DIMM.
Multiple windows allow multiple concurrent accesses, much like
tagged-command-queuing, and would likely be used by different threads or
different CPUs.
The NFIT specification defines a standard format for a BLK-aperture, but
the spec also allows for vendor specific layouts, and non-NFIT BLK
implementations may other designs for BLK I/O. For this reason "nd_blk"
calls back into platform-specific code to perform the I/O. One such
implementation is defined in the "Driver Writer's Guide" an "DSM
Interface Example".
Why BLK?
--------
While PMEM provides direct byte-addressable CPU-load/store access to
NVDIMM storage, it does not provide the best system RAS (recovery,
availability, and serviceability) model. An access to a corrupted
system-physical-address address causes a cpu exception while an access
to a corrupted address through an BLK-aperture causes that block window
to raise an error status in a register. The latter is more aligned with
the standard error model that host-bus-adapter attached disks present.
Also, if an administrator ever wants to replace a memory it is easier to
service a system at DIMM module boundaries. Compare this to PMEM where
data could be interleaved in an opaque hardware specific manner across
several DIMMs.
PMEM vs BLK
BLK-apertures solve this RAS problem, but their presence is also the
major contributing factor to the complexity of the ND subsystem. They
complicate the implementation because PMEM and BLK alias in DPA space.
Any given DIMM's DPA-range may contribute to one or more
system-physical-address sets of interleaved DIMMs, *and* may also be
accessed in its entirety through its BLK-aperture. Accessing a DPA
through a system-physical-address while simultaneously accessing the
same DPA through a BLK-aperture has undefined results. For this reason,
DIMM's with this dual interface configuration include a DSM function to
store/retrieve a LABEL. The LABEL effectively partitions the DPA-space
into exclusive system-physical-address and BLK-aperture accessible
regions. For simplicity a DIMM is allowed a PMEM "region" per each
interleave set in which it is a member. The remaining DPA space can be
carved into an arbitrary number of BLK devices with discontiguous
extents.
BLK-REGIONs, PMEM-REGIONs, Atomic Sectors, and DAX
--------------------------------------------------
One of the few
reasons to allow multiple BLK namespaces per REGION is so that each
BLK-namespace can be configured with a BTT with unique atomic sector
sizes. While a PMEM device can host a BTT the LABEL specification does
not provide for a sector size to be specified for a PMEM namespace.
This is due to the expectation that the primary usage model for PMEM is
via DAX, and the BTT is incompatible with DAX. However, for the cases
where an application or filesystem still needs atomic sector update
guarantees it can register a BTT on a PMEM device or partition. See
LIBND/NDCTL: Block Translation Table "btt"
Example NVDIMM Platform
-----------------------
For the remainder of this document the following diagram will be
referenced for any example sysfs layouts.
(a) (b) DIMM BLK-REGION
+-------------------+--------+--------+--------+
+------+ | pm0.0 | blk2.0 | pm1.0 | blk2.1 | 0 region2
| imc0 +--+- - - region0- - - +--------+ +--------+
+--+---+ | pm0.0 | blk3.0 | pm1.0 | blk3.1 | 1 region3
| +-------------------+--------v v--------+
+--+---+ | |
| cpu0 | region1
+--+---+ | |
| +----------------------------^ ^--------+
+--+---+ | blk4.0 | pm1.0 | blk4.0 | 2 region4
| imc1 +--+----------------------------| +--------+
+------+ | blk5.0 | pm1.0 | blk5.0 | 3 region5
+----------------------------+--------+--------+
In this platform we have four DIMMs and two memory controllers in one
socket. Each unique interface (BLK or PMEM) to DPA space is identified
by a region device with a dynamically assigned id (REGION0 - REGION5).
1. The first portion of DIMM0 and DIMM1 are interleaved as REGION0. A
single PMEM namespace is created in the REGION0-SPA-range that spans
DIMM0 and DIMM1 with a user-specified name of "pm0.0". Some of that
interleaved system-physical-address range is reclaimed as BLK-aperture
accessed space starting at DPA-offset (a) into each DIMM. In that
reclaimed space we create two BLK-aperture "namespaces" from REGION2 and
REGION3 where "blk2.0" and "blk3.0" are just human readable names that
could be set to any user-desired name in the LABEL.
2. In the last portion of DIMM0 and DIMM1 we have an interleaved
system-physical-address range, REGION1, that spans those two DIMMs as
well as DIMM2 and DIMM3. Some of REGION1 allocated to a PMEM namespace
named "pm1.0" the rest is reclaimed in 4 BLK-aperture namespaces (for
each DIMM in the interleave set), "blk2.1", "blk3.1", "blk4.0", and
"blk5.0".
3. The portion of DIMM2 and DIMM3 that do not participate in the REGION1
interleaved system-physical-address range (i.e. the DPA address below
offset (b) are also included in the "blk4.0" and "blk5.0" namespaces.
Note, that this example shows that BLK-aperture namespaces don't need to
be contiguous in DPA-space.
This bus is provided by the kernel under the device
/sys/devices/platform/nfit_test.0 when CONFIG_NFIT_TEST is enabled and
the nfit_test.ko module is loaded. This not only test LIBND but the
acpi_nfit.ko driver as well.
LIBND Kernel Device Model and LIBNDCTL Userspace API
----------------------------------------------------
What follows is a description of the LIBND sysfs layout and a
corresponding object hierarchy diagram as viewed through the LIBNDCTL
api. The example sysfs paths and diagrams are relative to the Example
NVDIMM Platform which is also the libnd bus used in the libndctl unit
test.
LIBNDCTL: Context
Every api call in the LIBNDCTL library requires a context that holds the
logging parameters and other library instance state. The library is
based on the libabc template:
https://git.kernel.org/cgit/linux/kernel/git/kay/libabc.git/ libndctl:
instantiate a new library context example
struct ndctl_ctx *ctx;
if (ndctl_new(&ctx) == 0)
return ctx;
else
return NULL;
LIBND/LIBNDCTL: Bus
-------------------
A bus has a 1:1 relationship with an NFIT. The current expectation for
ACPI based systems is that there is only ever one platform-global NFIT.
That said, it is trivial to register multiple NFITs, the specification
does not preclude it. The infrastructure supports multiple busses and
we we use this capability to test multiple NFIT configurations in the
unit test.
libnd: control class device in /sys/class
This character device accepts DSM messages to be passed to DIMM
identified by its NFIT handle.
/sys/class/nd/ndctl0
|-- dev
|-- device -> ../../../ndbus0
|-- subsystem -> ../../../../../../../class/nd
libnd: bus
struct nd_bus *nd_bus_register(struct device *parent,
struct nd_bus_descriptor *nfit_desc);
/sys/devices/platform/nfit_test.0/ndbus0
|-- btt0
|-- btt_seed
|-- commands
|-- nd
|-- nfit
|-- nmem0
|-- nmem1
|-- nmem2
|-- nmem3
|-- power
|-- provider
|-- region0
|-- region1
|-- region2
|-- region3
|-- region4
|-- region5
|-- uevent
`-- wait_probe
libndctl: bus enumeration example
Find the bus handle that describes the bus from Example NVDIMM Platform
static struct ndctl_bus *get_bus_by_provider(struct ndctl_ctx *ctx,
const char *provider)
{
struct ndctl_bus *bus;
ndctl_bus_foreach(ctx, bus)
if (strcmp(provider, ndctl_bus_get_provider(bus)) == 0)
return bus;
return NULL;
}
bus = get_bus_by_provider(ctx, "nfit_test.0");
LIBND/LIBNDCTL: DIMM (NMEM)
---------------------------
The DIMM device provides a character device for sending commands to
hardware, and it is a container for LABELs. If the DIMM is defined by
NFIT then an optional 'nfit' attribute sub-directory is available to add
NFIT-specifics.
Note that the kernel device name for "DIMMs" is "nmemX". The NFIT
describes these devices via "Memory Device to System Physical Address
Range Mapping Structure", and there is no requirement that they actually
be physical DIMMs, so we use a more generic name.
libnd: DIMM (NMEM)
struct nd_dimm *nd_dimm_create(struct nd_bus *nd_bus, void *provider_data,
const struct attribute_group **groups, unsigned long flags,
unsigned long *dsm_mask);
/sys/devices/platform/nfit_test.0/ndbus0
|-- nmem0
| |-- available_slots
| |-- commands
| |-- dev
| |-- devtype
| |-- driver -> ../../../../../bus/nd/drivers/nd_dimm
| |-- modalias
| |-- nfit
| | |-- device
| | |-- format
| | |-- handle
| | |-- phys_id
| | |-- rev_id
| | |-- serial
| | `-- vendor
| |-- state
| |-- subsystem -> ../../../../../bus/nd
| `-- uevent
|-- nmem1
[..]
libndctl: DIMM enumeration example
Note, in this example we are assuming NFIT-defined DIMMs which are
identified by an "nfit_handle" a 32-bit value where:
Bit 3:0 DIMM number within the memory channel
Bit 7:4 memory channel number
Bit 11:8 memory controller ID
Bit 15:12 socket ID (within scope of a Node controller if node controller is present)
Bit 27:16 Node Controller ID
Bit 31:28 Reserved
static struct ndctl_dimm *get_dimm_by_handle(struct ndctl_bus *bus,
unsigned int handle)
{
struct ndctl_dimm *dimm;
ndctl_dimm_foreach(bus, dimm)
if (ndctl_dimm_get_handle(dimm) == handle)
return dimm;
return NULL;
}
#define DIMM_HANDLE(n, s, i, c, d) \
(((n & 0xfff) << 16) | ((s & 0xf) << 12) | ((i & 0xf) << 8) \
| ((c & 0xf) << 4) | (d & 0xf))
dimm = get_dimm_by_handle(bus, DIMM_HANDLE(0, 0, 0, 0, 0));
LIBND/LIBNDCTL: Region
----------------------
A generic REGION device is registered for each PMEM range orBLK-aperture
set. Per the example there are 6 regions: 2 PMEM and 4 BLK-aperture
sets on the "nfit_test.0" bus. The primary role of regions are to be a
container of "mappings". A mapping is a tuple of <DIMM,
DPA-start-offset, length>.
LIBND provides a built-in driver for these REGION devices. This driver
is responsible for reconciling the aliased DPA mappings across all
regions, parsing the LABEL, if present, and then emitting NAMESPACE
devices with the resolved/exclusive DPA-boundaries for the nd_pmem or
nd_blk device driver to consume.
In addition to the generic attributes of "mapping"s, "interleave_ways"
and "size" the REGION device also exports some convenience attributes.
"nstype" indicates the integer type of namespace-device this region
emits, "devtype" duplicates the DEVTYPE variable stored by udev at the
'add' event, "modalias" duplicates the MODALIAS variable stored by udev
at the 'add' event, and finally, the optional "spa_index" is provided in
the case where the region is defined by a SPA.
libnd: region
struct nd_region *nd_pmem_region_create(struct nd_bus *nd_bus,
struct nd_region_desc *ndr_desc);
struct nd_region *nd_blk_region_create(struct nd_bus *nd_bus,
struct nd_region_desc *ndr_desc);
/sys/devices/platform/nfit_test.0/ndbus0
|-- region0
| |-- available_size
| |-- devtype
| |-- driver -> ../../../../../bus/nd/drivers/nd_region
| |-- init_namespaces
| |-- mapping0
| |-- mapping1
| |-- mappings
| |-- modalias
| |-- namespace0.0
| |-- namespace_seed
| |-- nfit
| | `-- spa_index
| |-- nstype
| |-- set_cookie
| |-- size
| |-- subsystem -> ../../../../../bus/nd
| `-- uevent
|-- region1
[..]
libndctl: region enumeration example
Sample region retrieval routines based on NFIT-unique data like
"spa_index" (interleave set id) for PMEM and "nfit_handle" (dimm id) for
BLK.
static struct ndctl_region *get_pmem_region_by_spa_index(struct ndctl_bus *bus,
unsigned int spa_index)
{
struct ndctl_region *region;
ndctl_region_foreach(bus, region) {
if (ndctl_region_get_type(region) != ND_DEVICE_REGION_PMEM)
continue;
if (ndctl_region_get_spa_index(region) == spa_index)
return region;
}
return NULL;
}
static struct ndctl_region *get_blk_region_by_dimm_handle(struct ndctl_bus *bus,
unsigned int handle)
{
struct ndctl_region *region;
ndctl_region_foreach(bus, region) {
struct ndctl_mapping *map;
if (ndctl_region_get_type(region) != ND_DEVICE_REGION_BLOCK)
continue;
ndctl_mapping_foreach(region, map) {
struct ndctl_dimm *dimm = ndctl_mapping_get_dimm(map);
if (ndctl_dimm_get_handle(dimm) == handle)
return region;
}
}
return NULL;
}
Why Not Encode the Region Type into the Region Name?
----------------------------------------------------
At first glance it seems since NFIT defines just PMEM and BLK interface
types that we should simply name REGION devices with something derived
from those type names. However, the ND subsystem explicitly keeps the
REGION name generic and expects userspace to always consider the
region-attributes for 4 reasons:
1. There are already more than two REGION and "namespace" types. For
PMEM there are two subtypes. As mentioned previously we have PMEM where
the constituent DIMM devices are known and anonymous PMEM. For BLK
regions the NFIT specification already anticipates vendor specific
implementations. The exact distinction of what a region contains is in
the region-attributes not the region-name or the region-devtype.
2. A region with zero child-namespaces is a possible configuration. For
example, the NFIT allows for a DCR to be published without a
corresponding BLK-aperture. This equates to a DIMM that can only accept
control/configuration messages, but no i/o through a descendant block
device. Again, this "type" is advertised in the attributes ('mappings'
== 0) and the name does not tell you much.
3. What if a third major interface type arises in the future? Outside
of vendor specific implementations, it's not difficult to envision a
third class of interface type beyond BLK and PMEM. With a generic name
for the REGION level of the device-hierarchy old userspace
implementations can still make sense of new kernel advertised
region-types. Userspace can always rely on the generic region
attributes like "mappings", "size", etc and the expected child devices
named "namespace". This generic format of the device-model hierarchy
allows the LIBND and LIBNDCTL implementations to be more uniform and
future-proof.
4. There are more robust mechanisms for determining the major type of a
region than a device name. See the next section, How Do I Determine the
Major Type of a Region?
How Do I Determine the Major Type of a Region?
----------------------------------------------
Outside of the blanket recommendation of "use libndctl", or simply
looking at the kernel header (/usr/include/linux/ndctl.h) to decode the
"nstype" integer attribute, here are some other options.
1. module alias lookup:
The whole point of region/namespace device type differentiation is to
decide which block-device driver will attach to a given LIBND namespace.
One can simply use the modalias to lookup the resulting module. It's
important to note that this method is robust in the presence of a
vendor-specific driver down the road. If a vendor-specific
implementation wants to supplant the standard nd_blk driver it can with
minimal impact to the rest of LIBND.
In fact, a vendor may also want to have a vendor-specific region-driver
(outside of nd_region). For example, if a vendor defined its own LABEL
format it would need its own region driver to parse that LABEL and emit
the resulting namespaces. The output from module resolution is more
accurate than a region-name or region-devtype.
2. udev:
The kernel "devtype" is registered in the udev database
# udevadm info --path=/devices/platform/nfit_test.0/ndbus0/region0
P: /devices/platform/nfit_test.0/ndbus0/region0
E: DEVPATH=/devices/platform/nfit_test.0/ndbus0/region0
E: DEVTYPE=nd_pmem
E: MODALIAS=nd:t2
E: SUBSYSTEM=nd
# udevadm info --path=/devices/platform/nfit_test.0/ndbus0/region4
P: /devices/platform/nfit_test.0/ndbus0/region4
E: DEVPATH=/devices/platform/nfit_test.0/ndbus0/region4
E: DEVTYPE=nd_blk
E: MODALIAS=nd:t3
E: SUBSYSTEM=nd
...and is available as a region attribute, but keep in mind that the
"devtype" does not indicate sub-type variations and scripts should
really be understanding the other attributes.
3. type specific attributes:
As it currently stands a BLK-aperture region will never have a
"nfit/spa_index" attribute, but neither will a non-NFIT PMEM region. A
BLK region with a "mappings" value of 0 is, as mentioned above, a DIMM
that does not allow I/O. A PMEM region with a "mappings" value of zero
is a simple system-physical-address range.
LIBND/LIBNDCTL: Namespace
-------------------------
A REGION, after resolving DPA aliasing and LABEL specified boundaries,
surfaces one or more "namespace" devices. The arrival of a "namespace"
device currently triggers either the nd_blk or nd_pmem driver to load
and register a disk/block device.
libnd: namespace
Here is a sample layout from the three major types of NAMESPACE where
namespace0.0 represents DIMM-info-backed PMEM (note that it has a 'uuid'
attribute), namespace2.0 represents a BLK namespace (note it has a
'sector_size' attribute) that, and namespace6.0 represents an anonymous
PMEM namespace (note that has no 'uuid' attribute due to not support a
LABEL).
/sys/devices/platform/nfit_test.0/ndbus0/region0/namespace0.0
|-- alt_name
|-- devtype
|-- dpa_extents
|-- modalias
|-- resource
|-- size
|-- subsystem -> ../../../../../../bus/nd
|-- type
|-- uevent
`-- uuid
/sys/devices/platform/nfit_test.0/ndbus0/region2/namespace2.0
|-- alt_name
|-- devtype
|-- dpa_extents
|-- modalias
|-- sector_size
|-- size
|-- subsystem -> ../../../../../../bus/nd
|-- type
|-- uevent
`-- uuid
/sys/devices/platform/nfit_test.1/ndbus1/region6/namespace6.0
|-- block
| `-- pmem0
|-- devtype
|-- driver -> ../../../../../../bus/nd/drivers/pmem
|-- modalias
|-- resource
|-- size
|-- subsystem -> ../../../../../../bus/nd
|-- type
`-- uevent
libndctl: namespace enumeration example
Namespaces are indexed relative to their parent region, example below. These indexes are mostly static from boot to boot, but subsystem makes no guarantees in this regard. For a static namespace identifier use its 'uuid' attribute.
static struct ndctl_namespace *get_namespace_by_id(struct ndctl_region *region,
unsigned int id)
{
struct ndctl_namespace *ndns;
ndctl_namespace_foreach(region, ndns)
if (ndctl_namespace_get_id(ndns) == id)
return ndns;
return NULL;
}
libndctl: namespace creation example
Idle namespaces are automatically created by the kernel if a given region has enough available capacity to create a new namespace. Namespace instantiation involves finding an idle namespace and configuring it. For the most part the setting of namespace attributes can occur in any order, the only constraint is that 'uuid' must be set before 'size'. This enables the kernel to track DPA allocations internally with a static identifier.
static int configure_namespace(struct ndctl_region *region,
struct ndctl_namespace *ndns,
struct namespace_parameters *parameters)
{
char devname[50];
snprintf(devname, sizeof(devname), "namespace%d.%d",
ndctl_region_get_id(region), paramaters->id);
ndctl_namespace_set_alt_name(ndns, devname);
/* 'uuid' must be set prior to setting size! */
ndctl_namespace_set_uuid(ndns, paramaters->uuid);
ndctl_namespace_set_size(ndns, paramaters->size);
/* unlike pmem namespaces, blk namespaces have a sector size */
if (parameters->lbasize)
ndctl_namespace_set_sector_size(ndns, parameters->lbasize);
ndctl_namespace_enable(ndns);
}
Why the Term "namespace"?
1. Why not "volume" for instance? "volume" ran the risk of confusing ND
as a volume manager like device-mapper.
2. The term originated to describe the sub-devices that can be created
within a NVME controller (see the nvme specification:
http://www.nvmexpress.org/specifications/), and NFIT namespaces are
meant to parallel the capabilities and configurability of
NVME-namespaces.
LIBND/LIBNDCTL: Block Translation Table "btt"
---------------------------------------------
A BTT (design document: http://pmem.io/2014/09/23/btt.html) is a stacked
block device driver that fronts either the whole block device or a
partition of a block device emitted by either a PMEM or BLK NAMESPACE.
libnd: btt layout
Every bus will start out with at least one BTT device which is the seed
device. To activate it set the "backing_dev", "uuid", and "sector_size"
attributes and then bind the device to the nd_btt driver.
/sys/devices/platform/nfit_test.1/ndbus0/btt0/
|-- backing_dev
|-- delete
|-- devtype
|-- modalias
|-- sector_size
|-- subsystem -> ../../../../../bus/nd
|-- uevent
`-- uuid
libndctl: btt creation example
Similar to namespaces an idle BTT device is automatically created per
bus. Each time this "seed" btt device is configured and enabled a new
seed is created. Creating a BTT configuration involves two steps of
finding and idle BTT and assigning it to front a PMEM or BLK namespace.
static struct ndctl_btt *get_idle_btt(struct ndctl_bus *bus)
{
struct ndctl_btt *btt;
ndctl_btt_foreach(bus, btt)
if (!ndctl_btt_is_enabled(btt) && !ndctl_btt_is_configured(btt))
return btt;
return NULL;
}
static int configure_btt(struct ndctl_bus *bus, struct btt_parameters *parameters)
{
btt = get_idle_btt(bus);
sprintf(bdevpath, "/dev/%s",
ndctl_namespace_get_block_device(parameters->ndns));
ndctl_btt_set_uuid(btt, parameters->uuid);
ndctl_btt_set_sector_size(btt, parameters->sector_size);
ndctl_btt_set_backing_dev(btt, parametes->bdevpath);
ndctl_btt_enable(btt);
}
Once instantiated a "nd_btt" link will be created under the
"backing_dev" (pmem0) block device:
/sys/block/pmem0/
|-- alignment_offset
|-- bdi -> ../../../../../../../virtual/bdi/259:0
|-- capability
|-- dev
|-- device -> ../../../namespace0.0
|-- discard_alignment
|-- ext_range
|-- holders
|-- inflight
|-- nd_btt -> ../../../../btt0
...and a new inactive seed device will appear on the bus.
Once a "backing_dev" is disabled its associated BTT will be
automatically deleted. This deletion is only at the device model level.
In order to destroy a BTT the "info block" needs to be destroyed.
Summary LIBNDCTL Diagram
------------------------
For the given example above, here is the view of the objects as seen by the LIBNDCTL api:
+---+
|CTX| +---------+ +--------------+ +---------------+
+-+-+ +-> REGION0 +---> NAMESPACE0.0 +--> PMEM8 "pm0.0" |
| | +---------+ +--------------+ +---------------+
+-------+ | | +---------+ +--------------+ +---------------+
| DIMM0 <-+ | +-> REGION1 +---> NAMESPACE1.0 +--> PMEM6 "pm1.0" |
+-------+ | | | +---------+ +--------------+ +---------------+
| DIMM1 <-+ +-v--+ | +---------+ +--------------+ +---------------+
+-------+ +-+BUS0+---> REGION2 +-+-> NAMESPACE2.0 +--> ND6 "blk2.0" |
| DIMM2 <-+ +----+ | +---------+ | +--------------+ +----------------------+
+-------+ | | +-> NAMESPACE2.1 +--> ND5 "blk2.1" | BTT2 |
| DIMM3 <-+ | +--------------+ +----------------------+
+-------+ | +---------+ +--------------+ +---------------+
+-> REGION3 +-+-> NAMESPACE3.0 +--> ND4 "blk3.0" |
| +---------+ | +--------------+ +----------------------+
| +-> NAMESPACE3.1 +--> ND3 "blk3.1" | BTT1 |
| +--------------+ +----------------------+
| +---------+ +--------------+ +---------------+
+-> REGION4 +---> NAMESPACE4.0 +--> ND2 "blk4.0" |
| +---------+ +--------------+ +---------------+
| +---------+ +--------------+ +----------------------+
+-> REGION5 +---> NAMESPACE5.0 +--> ND1 "blk5.0" | BTT0 |
+---------+ +--------------+ +---------------+------+
????{.n?+???????+%?????ݶ??w??{.n?+????{??G?????{ay? ʇڙ?,j??f???h?????????z_??(?階?ݢj"???m??????G????????????&???~???iO???z??v?^?m???? ????????I?
On Mon, May 4, 2015 at 1:26 PM, Toshi Kani <[email protected]> wrote:
> On Tue, 2015-04-28 at 14:24 -0400, Dan Williams wrote:
> :
>> +
>> +static int nd_acpi_register_region(struct acpi_nfit_desc *acpi_desc,
>> + struct nfit_spa *nfit_spa)
>> +{
>> + static struct nd_mapping nd_mappings[ND_MAX_MAPPINGS];
>> + struct acpi_nfit_spa *spa = nfit_spa->spa;
>> + struct nfit_memdev *nfit_memdev;
>> + struct nd_region_desc ndr_desc;
>> + int spa_type, count = 0;
>> + struct resource res;
>> + u16 spa_index;
>> +
>> + spa_type = nfit_spa_type(spa);
>> + spa_index = spa->spa_index;
>> + if (spa_index == 0) {
>> + dev_dbg(acpi_desc->dev, "%s: detected invalid spa index\n",
>> + __func__);
>> + return 0;
>> + }
>> +
>> + memset(&res, 0, sizeof(res));
>> + memset(&nd_mappings, 0, sizeof(nd_mappings));
>> + memset(&ndr_desc, 0, sizeof(ndr_desc));
>> + res.start = spa->spa_base;
>> + res.end = res.start + spa->spa_length - 1;
>> + ndr_desc.res = &res;
>> + ndr_desc.provider_data = nfit_spa;
>> + ndr_desc.attr_groups = nd_acpi_region_attribute_groups;
>> + list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) {
>> + struct acpi_nfit_memdev *memdev = nfit_memdev->memdev;
>> + struct nd_mapping *nd_mapping;
>> + struct nd_dimm *nd_dimm;
>> +
>> + if (memdev->spa_index != spa_index)
>> + continue;
>
> The libnd does not support memdev->flags, which contains "Memory Device
> State Flags" defined in Table 5-129 of ACPI 6.0. In case of major
> errors, we should only allow a failed NVDIMM be accessed with read-only
> for possible data recovery (or not allow any access when the data is
> completely lost), and should not let users operate normally over the
> corrupted data until the error is dealt properly.
I agree with setting read-only access when these flags show that the
battery is not ready to persist new writes, but I don't think we
should block access in the case where the restore from flash failed.
If the data is potentially corrupted we should log that fact, but
otherwise enable access. I.e. potentially corrupt data is better than
unavailable data. It's up to filesystem or application to maintain
its own checksums to catch data corruption.
> Can you set memdev->flags to nd_region(_desc) so that the pmem driver
> can check the status in nd_pmem_probe()? nd_pmem_probe() can then set
> the disk read-only or fail probing, and log errors accordingly.
Will do.
On Tue, 2015-04-28 at 14:25 -0400, Dan Williams wrote:
> Block devices from an nd bus, in addition to accepting "struct bio"
> based requests, also have the capability to perform byte-aligned
> accesses. By default only the bio/block interface is used. However, if
> another driver can make effective use of the byte-aligned capability it
> can claim/disable the block interface and use the byte-aligned "nd_io"
> interface.
>
> The BTT driver is the intended first consumer of this mechanism to allow
> layering atomic sector update guarantees on top of nd_io capable
> nd-bus-block-devices.
:
> +static int nd_btt_autodetect(struct nd_bus *nd_bus, struct nd_io *ndio,
> + struct block_device *bdev)
> +{
> + char name[BDEVNAME_SIZE];
> + struct nd_btt *nd_btt;
> + struct btt_sb *btt_sb;
> + u64 offset, checksum;
> + u32 lbasize;
> + u8 *uuid;
> + int rc;
> +
> + btt_sb = kzalloc(sizeof(*btt_sb), GFP_KERNEL);
> + if (!btt_sb)
> + return -ENODEV;
> +
> + offset = nd_partition_offset(bdev);
> + rc = ndio->rw_bytes(ndio, btt_sb, offset + SZ_4K, sizeof(*btt_sb), READ);
> + if (rc)
> + goto out_free_sb;
> +
> + if (get_capacity(bdev->bd_disk) < SZ_16M / 512)
> + goto out_free_sb;
> +
> + if (memcmp(btt_sb->signature, BTT_SIG, BTT_SIG_LEN) != 0)
> + goto out_free_sb;
> +
> + checksum = le64_to_cpu(btt_sb->checksum);
> + btt_sb->checksum = 0;
> + if (checksum != nd_btt_sb_checksum(btt_sb))
> + goto out_free_sb;
> + btt_sb->checksum = cpu_to_le64(checksum);
> +
> + uuid = kmemdup(btt_sb->uuid, 16, GFP_KERNEL);
> + if (!uuid)
> + goto out_free_sb;
> +
> + lbasize = le32_to_cpu(btt_sb->external_lbasize);
> + nd_btt = __nd_btt_create(nd_bus, lbasize, uuid);
When BTT is first set up, user binds a seed "btt0" to a block device,
such as /dev/pmem0. It then creates /dev/nd0 bound to /dev/pmem0.
After a reboot, nd_btt_autodetect() detects the BTT setup and creates a
new "btt1" since it is called after a seed "btt0" is created.
Therefore, it creates /dev/nd1 bound to /dev/pmem0 this time.
Is this how it is intended to work, i.e. "btt0" as the default seed btt?
While user should not rely on the name of /dev/nd%d, I thought this
device name change was confusing...
Thanks,
-Toshi
On Tue, May 12, 2015 at 9:33 AM, Toshi Kani <[email protected]> wrote:
> On Tue, 2015-04-28 at 14:25 -0400, Dan Williams wrote:
>> Block devices from an nd bus, in addition to accepting "struct bio"
>> based requests, also have the capability to perform byte-aligned
>> accesses. By default only the bio/block interface is used. However, if
>> another driver can make effective use of the byte-aligned capability it
>> can claim/disable the block interface and use the byte-aligned "nd_io"
>> interface.
>>
>> The BTT driver is the intended first consumer of this mechanism to allow
>> layering atomic sector update guarantees on top of nd_io capable
>> nd-bus-block-devices.
> :
>> +static int nd_btt_autodetect(struct nd_bus *nd_bus, struct nd_io *ndio,
>> + struct block_device *bdev)
>> +{
>> + char name[BDEVNAME_SIZE];
>> + struct nd_btt *nd_btt;
>> + struct btt_sb *btt_sb;
>> + u64 offset, checksum;
>> + u32 lbasize;
>> + u8 *uuid;
>> + int rc;
>> +
>> + btt_sb = kzalloc(sizeof(*btt_sb), GFP_KERNEL);
>> + if (!btt_sb)
>> + return -ENODEV;
>> +
>> + offset = nd_partition_offset(bdev);
>> + rc = ndio->rw_bytes(ndio, btt_sb, offset + SZ_4K, sizeof(*btt_sb), READ);
>> + if (rc)
>> + goto out_free_sb;
>> +
>> + if (get_capacity(bdev->bd_disk) < SZ_16M / 512)
>> + goto out_free_sb;
>> +
>> + if (memcmp(btt_sb->signature, BTT_SIG, BTT_SIG_LEN) != 0)
>> + goto out_free_sb;
>> +
>> + checksum = le64_to_cpu(btt_sb->checksum);
>> + btt_sb->checksum = 0;
>> + if (checksum != nd_btt_sb_checksum(btt_sb))
>> + goto out_free_sb;
>> + btt_sb->checksum = cpu_to_le64(checksum);
>> +
>> + uuid = kmemdup(btt_sb->uuid, 16, GFP_KERNEL);
>> + if (!uuid)
>> + goto out_free_sb;
>> +
>> + lbasize = le32_to_cpu(btt_sb->external_lbasize);
>> + nd_btt = __nd_btt_create(nd_bus, lbasize, uuid);
>
> When BTT is first set up, user binds a seed "btt0" to a block device,
> such as /dev/pmem0. It then creates /dev/nd0 bound to /dev/pmem0.
>
> After a reboot, nd_btt_autodetect() detects the BTT setup and creates a
> new "btt1" since it is called after a seed "btt0" is created.
> Therefore, it creates /dev/nd1 bound to /dev/pmem0 this time.
>
> Is this how it is intended to work, i.e. "btt0" as the default seed btt?
> While user should not rely on the name of /dev/nd%d, I thought this
> device name change was confusing...
So we can fix this to be at least as stable as the backing device
names [1], but as far as I can see we would need to start using the
backing device name in the btt device name. A strawman proposal is to
append 's' to indicated 'sectored'. So /dev/pmem0s is the btt
instance fronting /dev/pmem0. Other examples:
/dev/pmem0p1s
/dev/ndblk0.0s
/dev/ndblk0.0p1s
...
Thoughts?
[1]: https://lists.01.org/pipermail/linux-nvdimm/2015-April/000636.html
> -----Original Message-----
> From: Linux-nvdimm [mailto:[email protected]] On Behalf
> Of Dan Williams
> Sent: Thursday, May 14, 2015 7:42 PM
> To: Kani, Toshimitsu
> Cc: Neil Brown; Greg KH; [email protected]; linux-
> [email protected]
> Subject: Re: [Linux-nvdimm] [PATCH v2 18/20] libnd: infrastructure for
> btt devices
>
...
> So we can fix this to be at least as stable as the backing device
> names [1], but as far as I can see we would need to start using the
> backing device name in the btt device name. A strawman proposal is to
> append 's' to indicated 'sectored'. So /dev/pmem0s is the btt
> instance fronting /dev/pmem0. Other examples:
>
> /dev/pmem0p1s
> /dev/ndblk0.0s
> /dev/ndblk0.0p1s
> ...
>
> Thoughts?
>
> [1]: https://lists.01.org/pipermail/linux-nvdimm/2015-April/000636.html
I like that; it also hints to the user that another driver has already
claimed /dev/pmem0, similar to how the presence of /dev/sda1, /dev/sda2,
etc. hints that a program has partitioned /dev/sda.
Dan Williams <[email protected]> writes:
> @@ -919,10 +920,31 @@ static inline const char *e820_type_to_string(int e820_type)
> case E820_NVS: return "ACPI Non-volatile Storage";
> case E820_UNUSABLE: return "Unusable memory";
> case E820_PRAM: return "Persistent RAM";
> + case E820_PMEM: return "Persistent I/O Memory";
Minor nit, but nobody refers to this stuff as persistent I/O memory.
I've heard it referred to as persistent memory or storage class memory.
In the end, I doubt it matters.
Acked-by: Jeff Moyer <[email protected]>
On Fri, May 15, 2015 at 8:43 AM, Jeff Moyer <[email protected]> wrote:
> Dan Williams <[email protected]> writes:
>
>> @@ -919,10 +920,31 @@ static inline const char *e820_type_to_string(int e820_type)
>> case E820_NVS: return "ACPI Non-volatile Storage";
>> case E820_UNUSABLE: return "Unusable memory";
>> case E820_PRAM: return "Persistent RAM";
>> + case E820_PMEM: return "Persistent I/O Memory";
>
> Minor nit, but nobody refers to this stuff as persistent I/O memory.
> I've heard it referred to as persistent memory or storage class memory.
> In the end, I doubt it matters.
Had we not polluted the waters with E820_PRAM I would have chosen
"Persistent Memory", but as we now need to support both I'd assume
they need different names.
> Acked-by: Jeff Moyer <[email protected]>
Thanks Jeff.
Dan Williams <[email protected]> writes:
Looks like the Kconfig stuff has been worked out between you and Rafael,
so I won't comment on that.
> diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
> new file mode 100644
> index 000000000000..9f0b24390d1b
> --- /dev/null
> +++ b/drivers/block/nd/acpi.c
> @@ -0,0 +1,475 @@
> +/*
> + * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of version 2 of the GNU General Public License as
> + * published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope that it will be useful, but
> + * WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> + * General Public License for more details.
> + */
> +#include <linux/list_sort.h>
> +#include <linux/module.h>
> +#include <linux/list.h>
> +#include <linux/acpi.h>
> +#include "acpi_nfit.h"
> +#include "libnd.h"
> +
> +static bool warn_checksum;
> +module_param(warn_checksum, bool, S_IRUGO|S_IWUSR);
> +MODULE_PARM_DESC(warn_checksum, "Turn checksum errors into warnings");
Is this just a debugging option?
> +
> +enum {
> + NFIT_ACPI_NOTIFY_TABLE = 0x80,
> +};
This is unused by this patch.
The rest looks ok to me.
-Jeff
Dan Williams <[email protected]> writes:
> +config NFIT_TEST
> + tristate "NFIT TEST: Manufactured NFIT for interface testing"
> + depends on DMA_CMA
> + depends on LIBND=m
> + depends on ND_ACPI
> + depends on m
> + help
> + For development purposes register a manufactured
> + NFIT table to verify the resulting device model topology.
> + Note, this module arranges for ioremap_cache() to be
> + overridden locally to allow simulation of system-memory as an
> + io-memory-resource.
> +
> + Note, this test expects to be able to find at least 256MB of
> + CMA space (CONFIG_CMA_SIZE_MBYTES, cma=) or it will fail to
> + load.
> +
> + Say N unless you are doing development of the 'nd' subsystem.
> +
Too many TLAs. I'm guessing CMA means Conventional Memory Area to you.
To me it means contiguous memory allocator. Anyway, please define
acronyms when you use them, especially in help text. The help text also
doesn't really explain where it will find this memory. Would it be
possible to provide more direction there?
I don't have any useful commentary on the patch itself. I do wonder if
you shouldn't move this to the end, as it's hardly an integral part of
the patch set.
Cheers,
Jeff
On Fri, May 15, 2015 at 12:44 PM, Jeff Moyer <[email protected]> wrote:
> Dan Williams <[email protected]> writes:
>
> Looks like the Kconfig stuff has been worked out between you and Rafael,
> so I won't comment on that.
>
>> diff --git a/drivers/block/nd/acpi.c b/drivers/block/nd/acpi.c
>> new file mode 100644
>> index 000000000000..9f0b24390d1b
>> --- /dev/null
>> +++ b/drivers/block/nd/acpi.c
>> @@ -0,0 +1,475 @@
>> +/*
>> + * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
>> + *
>> + * This program is free software; you can redistribute it and/or modify
>> + * it under the terms of version 2 of the GNU General Public License as
>> + * published by the Free Software Foundation.
>> + *
>> + * This program is distributed in the hope that it will be useful, but
>> + * WITHOUT ANY WARRANTY; without even the implied warranty of
>> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
>> + * General Public License for more details.
>> + */
>> +#include <linux/list_sort.h>
>> +#include <linux/module.h>
>> +#include <linux/list.h>
>> +#include <linux/acpi.h>
>> +#include "acpi_nfit.h"
>> +#include "libnd.h"
>> +
>> +static bool warn_checksum;
>> +module_param(warn_checksum, bool, S_IRUGO|S_IWUSR);
>> +MODULE_PARM_DESC(warn_checksum, "Turn checksum errors into warnings");
>
> Is this just a debugging option?
Yes, but I've deleted it in the next rev of the code since the ACPI
core will have already done the checksum. The driver need not
implement it's own checksum.
>> +
>> +enum {
>> + NFIT_ACPI_NOTIFY_TABLE = 0x80,
>> +};
>
> This is unused by this patch.
Yes. I went ahead and deleted all the ACPI notification
infrastructure until we're ready to implement hot-add.
> The rest looks ok to me.
Thanks Jeff.
On Fri, May 15, 2015 at 1:25 PM, Jeff Moyer <[email protected]> wrote:
> Dan Williams <[email protected]> writes:
>
>> +config NFIT_TEST
>> + tristate "NFIT TEST: Manufactured NFIT for interface testing"
>> + depends on DMA_CMA
>> + depends on LIBND=m
>> + depends on ND_ACPI
>> + depends on m
>> + help
>> + For development purposes register a manufactured
>> + NFIT table to verify the resulting device model topology.
>> + Note, this module arranges for ioremap_cache() to be
>> + overridden locally to allow simulation of system-memory as an
>> + io-memory-resource.
>> +
>> + Note, this test expects to be able to find at least 256MB of
>> + CMA space (CONFIG_CMA_SIZE_MBYTES, cma=) or it will fail to
>> + load.
>> +
>> + Say N unless you are doing development of the 'nd' subsystem.
>> +
>
> Too many TLAs. I'm guessing CMA means Conventional Memory Area to you.
> To me it means contiguous memory allocator.
I means Contiguous Memory Allocator to me too.
> Anyway, please define
> acronyms when you use them, especially in help text. The help text also
> doesn't really explain where it will find this memory. Would it be
> possible to provide more direction there?
I didn't think I needed to define CMA in the context of Kconfig, but
I'll replace it with CONFIG_DMA_CMA to be more clear.
> I don't have any useful commentary on the patch itself. I do wonder if
> you shouldn't move this to the end, as it's hardly an integral part of
> the patch set.
True. The patches are currently in "development order" in that I
created the test infrastructure before the rest of the implementation.
But, I agree it makes sense to move this to the end for the next
posting.
Dan Williams <[email protected]> writes:
> This is the position (device topology) independent method to find all
> the libnd buses in the system. The expectation is that there will only
> ever be one "nd" bus discovered via /sys/class/nd/ndctl0. However, we
> allow for the possibility of multiple buses and they will listed in
> discovery order as ndctl0...ndctlN. This character device hosts the
> ioctl for passing control messages (inspired by the ACPI-NFIT DSM
> interface commands).
>
> Note, nd_ioctl() and the backing ->ndctl() implementation are defined in
> a subsequent patch.
Acked-by: Jeff Moyer <[email protected]>
> -----Original Message-----
> From: Linux-nvdimm [mailto:[email protected]] On Behalf Of
> Dan Williams
> Sent: Tuesday, April 28, 2015 1:26 PM
> To: [email protected]
> Cc: Ingo Molnar; Neil Brown; Greg KH; Dave Chinner; linux-
> [email protected]; Andy Lutomirski; Jens Axboe; H. Peter Anvin;
> Christoph Hellwig
> Subject: [Linux-nvdimm] [PATCH v2 19/20] nd_btt: atomic sector updates
>
> From: Vishal Verma <[email protected]>
>
> BTT stands for Block Translation Table, and is a way to provide power
> fail sector atomicity semantics for block devices that have the ability
> to perform byte granularity IO. It relies on the ->rw_bytes() capability
> of provided nd namespace devices.
>
> The BTT works as a stacked blocked device, and reserves a chunk of space
> from the backing device for its accounting metadata. BLK namespaces may
> mandate use of a BTT and expect the bus to initialize a BTT if not
> already present. Otherwise if a BTT is desired for other namespaces (or
> partitions of a namespace) a BTT may be manually configured.
...
Running btt above pmem with a variety of workloads, I see an awful lot
of time spent in two places:
* _raw_spin_lock
* btt_make_request
This occurs for fio to raw /dev/ndN devices, ddpt over ext4 or xfs,
cp -R of large directories, and running make on the linux kernel.
Some specific results:
fio 4 KiB random reads, WC cache type, memcpy:
* 43175 MB/s, 8 M IOPS pmem0 and pmem1
* 18500 MB/s, 1.5 M IOPS nd0 and nd1
fio 4 KiB random reads, WC cache type, memcpy with non-temporal
loads (when everything is 64-byte aligned):
* 33814 MB/s, 4.3 M IOPS nd0 and nd1
Zeroing out 32 MiB with ddpt:
* 19 s, 1800 MiB/s pmem
* 55 s, 625 MiB/s btt
If btt_make_request needs to stall this much, maybe it'd be better
to utilize the blk-mq request queues, keeping requests in per-CPU
queues while they're waiting, and using IPIs for completion
interrupts when they're finally done.
fio 4 KiB random reads without non-temporal memcpy
==================================================
perf top shows memcpy_erms taking all the time, a function that
uses 8-byte REP; MOVSB instructions:
85.78% [kernel] [k] memcpy_erms
1.21% [kernel] [k] _raw_spin_lock
0.72% [nd_btt] [k] btt_make_request
0.67% [kernel] [k] do_blockdev_direct_IO
0.47% fio [.] get_io_u
fio 4 KiB random reads with non-temporal memcpy
===============================================
perf top shows there are still quite a few unaligned accesses
resulting in legacy memcpy, but about equal time is now spent
in legacy vs NT memcpy:
30.47% [kernel] [k] memcpy_erms
26.27% [kernel] [k] memcpy_lnt_st_64
5.37% [kernel] [k] _raw_spin_lock
2.20% [kernel] [k] btt_make_request
2.03% [kernel] [k] do_blockdev_direct_IO
1.41% fio [.] get_io_u
1.22% [kernel] [k] btt_map_read
1.15% [kernel] [k] pmem_rw_bytes
1.01% [kernel] [k] nd_btt_rw_bytes
0.98% [kernel] [k] nd_region_acquire_lane
0.89% fio [.] get_next_rand_block
0.88% fio [.] thread_main
0.79% fio [.] ios_completed
0.76% fio [.] td_io_queue
0.75% [kernel] [k] _raw_spin_lock_irqsave
0.68% [kernel] [k] kmem_cache_free
0.66% [kernel] [k] kmem_cache_alloc
0.59% [kernel] [k] __audit_syscall_exit
0.57% [kernel] [k] aio_complete
0.54% [kernel] [k] do_io_submit
0.52% [kernel] [k] _raw_spin_unlock_irqrestore
fio randrw workload
===================
perf top shows that adding writes to the mix brings btt_make_request
its cpu_relax() loop to the forefront:
21.09% [nd_btt] [k] btt_make_request
19.06% [kernel] [k] memcpy_erms
14.35% [kernel] [k] _raw_spin_lock
10.38% [nd_pmem] [k] memcpy_lnt_st_64
1.57% [kernel] [k] do_blockdev_direct_IO
1.51% [nd_pmem] [k] memcpy_lt_snt_64
1.43% [nd_btt] [k] nd_btt_rw_bytes
1.39% [kernel] [k] radix_tree_next_chunk
1.33% [kernel] [k] put_page
1.21% [nd_pmem] [k] pmem_rw_bytes
1.11% fio [.] get_io_u
0.90% fio [.] io_u_queued_complete
0.74% [kernel] [k] system_call
0.72% [libnd] [k] nd_region_acquire_lane
0.71% [nd_btt] [k] btt_map_read
0.62% fio [.] thread_main
inside btt_make_request:
? /* Wait if the new block is being read from */
? for (i = 0; i < arena->nfree; i++)
2.98 ? ? je 2b4
0.05 ? mov 0x60(%r14),%rax
0.00 ? mov %ebx,%edx
? xor %esi,%esi
0.03 ? or $0x80000000,%edx
0.05 ? nop
? while (arena->rtt[i] == (RTT_VALID | new_postmap))
22.98 ?290: mov %esi,%edi
0.01 ? cmp %edx,(%rax,%rdi,4)
30.97 ? lea 0x0(,%rdi,4),%rcx
21.05 ? ? jne 2ab
? nop
? }
?
? /* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
? static inline void rep_nop(void)
? {
? asm volatile("rep; nop" ::: "memory");
?2a0: pause
? mov 0x60(%r14),%rax
? cmp (%rax,%rcx,1),%edx
? ? je 2a0
? }
ddpt zeroing out
================
perf top shows 27% in spinlocks, and 14% in btt_make_request (all in
the "wait if the new block is being read from" loop).
26.48% [kernel] [k] _raw_spin_lock
14.46% [nd_btt] [k] btt_make_request
13.14% [kernel] [k] memcpy_erms
10.34% [kernel] [k] copy_user_enhanced_fast_string
3.12% [nd_pmem] [k] memcpy_lt_snt_64
1.15% [kernel] [k] __block_commit_write.isra.21
0.96% [nd_pmem] [k] pmem_rw_bytes
0.96% [nd_btt] [k] nd_btt_rw_bytes
0.86% [kernel] [k] unlock_page
0.65% [kernel] [k] _raw_spin_lock_irqsave
0.58% [kernel] [k] bdev_read_only
0.56% [kernel] [k] release_pages
0.54% [nd_pmem] [k] memcpy_lnt_st_64
0.53% [ext4] [k] ext4_mark_iloc_dirty
0.52% [kernel] [k] __wake_up_bit
0.52% [kernel] [k] __clear_user
---
Robert Elliott, HP Server Storage
On Sat, May 16, 2015 at 6:19 PM, Elliott, Robert (Server Storage)
<[email protected]> wrote:
>
>> -----Original Message-----
>> From: Linux-nvdimm [mailto:[email protected]] On Behalf Of
>> Dan Williams
>> Sent: Tuesday, April 28, 2015 1:26 PM
>> To: [email protected]
>> Cc: Ingo Molnar; Neil Brown; Greg KH; Dave Chinner; linux-
>> [email protected]; Andy Lutomirski; Jens Axboe; H. Peter Anvin;
>> Christoph Hellwig
>> Subject: [Linux-nvdimm] [PATCH v2 19/20] nd_btt: atomic sector updates
>>
>> From: Vishal Verma <[email protected]>
>>
>> BTT stands for Block Translation Table, and is a way to provide power
>> fail sector atomicity semantics for block devices that have the ability
>> to perform byte granularity IO. It relies on the ->rw_bytes() capability
>> of provided nd namespace devices.
>>
>> The BTT works as a stacked blocked device, and reserves a chunk of space
>> from the backing device for its accounting metadata. BLK namespaces may
>> mandate use of a BTT and expect the bus to initialize a BTT if not
>> already present. Otherwise if a BTT is desired for other namespaces (or
>> partitions of a namespace) a BTT may be manually configured.
> ...
>
> Running btt above pmem with a variety of workloads, I see an awful lot
> of time spent in two places:
> * _raw_spin_lock
> * btt_make_request
>
> This occurs for fio to raw /dev/ndN devices, ddpt over ext4 or xfs,
> cp -R of large directories, and running make on the linux kernel.
>
> Some specific results:
>
> fio 4 KiB random reads, WC cache type, memcpy:
> * 43175 MB/s, 8 M IOPS pmem0 and pmem1
> * 18500 MB/s, 1.5 M IOPS nd0 and nd1
>
> fio 4 KiB random reads, WC cache type, memcpy with non-temporal
> loads (when everything is 64-byte aligned):
> * 33814 MB/s, 4.3 M IOPS nd0 and nd1
>
> Zeroing out 32 MiB with ddpt:
> * 19 s, 1800 MiB/s pmem
> * 55 s, 625 MiB/s btt
>
> If btt_make_request needs to stall this much, maybe it'd be better
> to utilize the blk-mq request queues, keeping requests in per-CPU
> queues while they're waiting, and using IPIs for completion
> interrupts when they're finally done.
2 items to check:
1/ make sure you have a your btt sector size set to 4k which cuts down
the overhead by a factor of 8.
2/ boot with nr_cpus=256 or lower.
Ross noticed that CONFIG_NR_CPUS is set quite high on distro kernels
which revealed that we should have been using nr_cpu_ids and percpu
variables for nd_region_acquire_lane() from the outset. This fix is
coming in v3.
On May 16, 2015 12:48 AM, "Dan Williams" <[email protected]> wrote:
>
> On Fri, May 15, 2015 at 8:43 AM, Jeff Moyer <[email protected]> wrote:
> > Dan Williams <[email protected]> writes:
> >
> >> @@ -919,10 +920,31 @@ static inline const char *e820_type_to_string(int e820_type)
> >> case E820_NVS: return "ACPI Non-volatile Storage";
> >> case E820_UNUSABLE: return "Unusable memory";
> >> case E820_PRAM: return "Persistent RAM";
> >> + case E820_PMEM: return "Persistent I/O Memory";
> >
> > Minor nit, but nobody refers to this stuff as persistent I/O memory.
> > I've heard it referred to as persistent memory or storage class memory.
> > In the end, I doubt it matters.
>
> Had we not polluted the waters with E820_PRAM I would have chosen
> "Persistent Memory", but as we now need to support both I'd assume
> they need different names.
We could reverse it and call E820_PRAM "legacy persistent RAM" or
something similar.
</bikeshed>
>
> > Acked-by: Jeff Moyer <[email protected]>
>
> Thanks Jeff.
On Sun, 2015-05-17 at 01:19 +0000, Elliott, Robert (Server Storage)
wrote:
> > -----Original Message-----
> > From: Linux-nvdimm [mailto:[email protected]] On Behalf Of
> > Dan Williams
> > Sent: Tuesday, April 28, 2015 1:26 PM
> > To: [email protected]
> > Cc: Ingo Molnar; Neil Brown; Greg KH; Dave Chinner; linux-
> > [email protected]; Andy Lutomirski; Jens Axboe; H. Peter Anvin;
> > Christoph Hellwig
> > Subject: [Linux-nvdimm] [PATCH v2 19/20] nd_btt: atomic sector updates
> >
> > From: Vishal Verma <[email protected]>
> >
> > BTT stands for Block Translation Table, and is a way to provide power
> > fail sector atomicity semantics for block devices that have the ability
> > to perform byte granularity IO. It relies on the ->rw_bytes() capability
> > of provided nd namespace devices.
> >
> > The BTT works as a stacked blocked device, and reserves a chunk of space
> > from the backing device for its accounting metadata. BLK namespaces may
> > mandate use of a BTT and expect the bus to initialize a BTT if not
> > already present. Otherwise if a BTT is desired for other namespaces (or
> > partitions of a namespace) a BTT may be manually configured.
> ...
>
> Running btt above pmem with a variety of workloads, I see an awful lot
> of time spent in two places:
> * _raw_spin_lock
> * btt_make_request
I'm curious about why _raw_spin_lock might be taking up so much time -
does the test system have more than 256 CPUs? Even ignoring Dan's fix
for nr_cpus which will allow the lockless path optimization to kick in
(coming in v3), the lane spinlocks should be largely uncontested unless
you have more than 256 CPUs.
One explanation could be that this is being dominated by 'map_locks' -
I'd need to get my perf setup going again to find out. (The more I think
about it, the more I'm convinced this is the case. Fortunately it is
simple to reduce map_lock contention by increasing the number of locks).
>
> This occurs for fio to raw /dev/ndN devices, ddpt over ext4 or xfs,
> cp -R of large directories, and running make on the linux kernel.
>
> Some specific results:
>
> fio 4 KiB random reads, WC cache type, memcpy:
> * 43175 MB/s, 8 M IOPS pmem0 and pmem1
> * 18500 MB/s, 1.5 M IOPS nd0 and nd1
>
> fio 4 KiB random reads, WC cache type, memcpy with non-temporal
> loads (when everything is 64-byte aligned):
> * 33814 MB/s, 4.3 M IOPS nd0 and nd1
>
> Zeroing out 32 MiB with ddpt:
> * 19 s, 1800 MiB/s pmem
> * 55 s, 625 MiB/s btt
>
> If btt_make_request needs to stall this much, maybe it'd be better
> to utilize the blk-mq request queues, keeping requests in per-CPU
> queues while they're waiting, and using IPIs for completion
> interrupts when they're finally done.
>
>
> fio 4 KiB random reads without non-temporal memcpy
> ==================================================
> perf top shows memcpy_erms taking all the time, a function that
> uses 8-byte REP; MOVSB instructions:
> 85.78% [kernel] [k] memcpy_erms
> 1.21% [kernel] [k] _raw_spin_lock
> 0.72% [nd_btt] [k] btt_make_request
> 0.67% [kernel] [k] do_blockdev_direct_IO
> 0.47% fio [.] get_io_u
>
> fio 4 KiB random reads with non-temporal memcpy
> ===============================================
> perf top shows there are still quite a few unaligned accesses
> resulting in legacy memcpy, but about equal time is now spent
> in legacy vs NT memcpy:
I take it memcpy_lnt_st_64 expects 64B aligned accesses? The BTT will
always have some unaligned accesses - the map updates are 4B long. Flog
updates are 16B long, but about half of them will not start at a 64B
alignment due to how it is laid out. I'm open to hearing ideas about how
this could be improved. One way, as Dan already said is reducing the
number of metadata updates by configuring a larger (4K) sector size.
> 30.47% [kernel] [k] memcpy_erms
> 26.27% [kernel] [k] memcpy_lnt_st_64
> 5.37% [kernel] [k] _raw_spin_lock
> 2.20% [kernel] [k] btt_make_request
> 2.03% [kernel] [k] do_blockdev_direct_IO
> 1.41% fio [.] get_io_u
> 1.22% [kernel] [k] btt_map_read
> 1.15% [kernel] [k] pmem_rw_bytes
> 1.01% [kernel] [k] nd_btt_rw_bytes
> 0.98% [kernel] [k] nd_region_acquire_lane
> 0.89% fio [.] get_next_rand_block
> 0.88% fio [.] thread_main
> 0.79% fio [.] ios_completed
> 0.76% fio [.] td_io_queue
> 0.75% [kernel] [k] _raw_spin_lock_irqsave
> 0.68% [kernel] [k] kmem_cache_free
> 0.66% [kernel] [k] kmem_cache_alloc
> 0.59% [kernel] [k] __audit_syscall_exit
> 0.57% [kernel] [k] aio_complete
> 0.54% [kernel] [k] do_io_submit
> 0.52% [kernel] [k] _raw_spin_unlock_irqrestore
>
> fio randrw workload
> ===================
> perf top shows that adding writes to the mix brings btt_make_request
> its cpu_relax() loop to the forefront:
This is good data - we've thought about benchmarking an alternate
strategy for the 'Read Tracking Table' - instead of a linear array, use
a hash table (hash new_postmap into an RTT slot) . It will add false
waiting due to hash collisions, but the writers won't have to walk the
array - they can just hash into the right spot.
Another option that Dan suggested could be instead of waiting in the
cpu_relax() loop, we drop the lane, and start over with _acquire_lane.
This way, we are waiting in the spin_lock stage, thus giving a chance to
other threads/CPUs that resolve to the same spinlock to run.
> 21.09% [nd_btt] [k] btt_make_request
> 19.06% [kernel] [k] memcpy_erms
> 14.35% [kernel] [k] _raw_spin_lock
> 10.38% [nd_pmem] [k] memcpy_lnt_st_64
> 1.57% [kernel] [k] do_blockdev_direct_IO
> 1.51% [nd_pmem] [k] memcpy_lt_snt_64
> 1.43% [nd_btt] [k] nd_btt_rw_bytes
> 1.39% [kernel] [k] radix_tree_next_chunk
> 1.33% [kernel] [k] put_page
> 1.21% [nd_pmem] [k] pmem_rw_bytes
> 1.11% fio [.] get_io_u
> 0.90% fio [.] io_u_queued_complete
> 0.74% [kernel] [k] system_call
> 0.72% [libnd] [k] nd_region_acquire_lane
> 0.71% [nd_btt] [k] btt_map_read
> 0.62% fio [.] thread_main
>
> inside btt_make_request:
>
> ¦ /* Wait if the new block is being read from */
> ¦ for (i = 0; i < arena->nfree; i++)
> 2.98 ¦ ? je 2b4
> 0.05 ¦ mov 0x60(%r14),%rax
> 0.00 ¦ mov %ebx,%edx
> ¦ xor %esi,%esi
> 0.03 ¦ or $0x80000000,%edx
> 0.05 ¦ nop
> ¦ while (arena->rtt[i] == (RTT_VALID | new_postmap))
> 22.98 ¦290: mov %esi,%edi
> 0.01 ¦ cmp %edx,(%rax,%rdi,4)
> 30.97 ¦ lea 0x0(,%rdi,4),%rcx
> 21.05 ¦ ? jne 2ab
> ¦ nop
> ¦ }
> ¦
> ¦ /* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
> ¦ static inline void rep_nop(void)
> ¦ {
> ¦ asm volatile("rep; nop" ::: "memory");
> ¦2a0: pause
> ¦ mov 0x60(%r14),%rax
> ¦ cmp (%rax,%rcx,1),%edx
> ¦ ? je 2a0
> ¦ }
>
>
> ddpt zeroing out
> ================
> perf top shows 27% in spinlocks, and 14% in btt_make_request (all in
> the "wait if the new block is being read from" loop).
>
> 26.48% [kernel] [k] _raw_spin_lock
> 14.46% [nd_btt] [k] btt_make_request
> 13.14% [kernel] [k] memcpy_erms
> 10.34% [kernel] [k] copy_user_enhanced_fast_string
> 3.12% [nd_pmem] [k] memcpy_lt_snt_64
> 1.15% [kernel] [k] __block_commit_write.isra.21
> 0.96% [nd_pmem] [k] pmem_rw_bytes
> 0.96% [nd_btt] [k] nd_btt_rw_bytes
> 0.86% [kernel] [k] unlock_page
> 0.65% [kernel] [k] _raw_spin_lock_irqsave
> 0.58% [kernel] [k] bdev_read_only
> 0.56% [kernel] [k] release_pages
> 0.54% [nd_pmem] [k] memcpy_lnt_st_64
> 0.53% [ext4] [k] ext4_mark_iloc_dirty
> 0.52% [kernel] [k] __wake_up_bit
> 0.52% [kernel] [k] __clear_user
>
> ---
> Robert Elliott, HP Server Storage
Thanks, this is great analysis - I'd be quite interested in how these
numbers look like with the two changes Dan mentioned.
-Vishal
????{.n?+???????+%?????ݶ??w??{.n?+????{??G?????{ay? ʇڙ?,j??f???h?????????z_??(?階?ݢj"???m??????G????????????&???~???iO???z??v?^?m???? ????????I?
> -----Original Message-----
> From: Linux-nvdimm [mailto:[email protected]] On Behalf
> Of Dan Williams
> Sent: Tuesday, April 28, 2015 1:25 PM
> To: [email protected]
> Cc: Neil Brown; Greg KH; [email protected]
> Subject: [Linux-nvdimm] [PATCH v2 07/20] libnd, nd_dimm: dimm driver and
> base libnd device-driver infrastructure
>
...
> diff --git a/drivers/block/nd/dimm.c b/drivers/block/nd/dimm.c
> new file mode 100644
> index 000000000000..a4c8e3ffe97c
> --- /dev/null
> +++ b/drivers/block/nd/dimm.c
> @@ -0,0 +1,93 @@
> +/*
> + * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of version 2 of the GNU General Public License as
> + * published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope that it will be useful, but
> + * WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> + * General Public License for more details.
> + */
> +#include <linux/vmalloc.h>
> +#include <linux/module.h>
> +#include <linux/device.h>
> +#include <linux/sizes.h>
> +#include <linux/ndctl.h>
> +#include <linux/slab.h>
> +#include <linux/mm.h>
> +#include <linux/nd.h>
> +#include "nd.h"
> +
> +static void free_data(struct nd_dimm_drvdata *ndd)
> +{
> + if (!ndd)
> + return;
> +
> + if (ndd->data && is_vmalloc_addr(ndd->data))
> + vfree(ndd->data);
> + else
> + kfree(ndd->data);
> + kfree(ndd);
> +}
> +
> +static int nd_dimm_probe(struct device *dev)
> +{
> + struct nd_dimm_drvdata *ndd;
> + int rc;
> +
> + ndd = kzalloc(sizeof(*ndd), GFP_KERNEL);
> + if (!ndd)
> + return -ENOMEM;
> +
> + dev_set_drvdata(dev, ndd);
> + ndd->dev = dev;
> +
> + rc = nd_dimm_init_nsarea(ndd);
> + if (rc)
> + goto err;
> +
> + rc = nd_dimm_init_config_data(ndd);
> + if (rc)
> + goto err;
> +
> + dev_dbg(dev, "config data size: %d\n", ndd->nsarea.config_size);
> +
> + return 0;
> +
> + err:
> + free_data(ndd);
> + return rc;
> +
> +}
> +
> +static int nd_dimm_remove(struct device *dev)
> +{
> + struct nd_dimm_drvdata *ndd = dev_get_drvdata(dev);
> +
> + free_data(ndd);
> +
> + return 0;
> +}
It would reduce the slight window for stale pointer usage if you add:
dev_set_drvdata(dev, NULL);
before
free_data(ndd);
in both nd_dimm_remove and the err exit for nd_dimm_probe.
The same comment applies to all the drivers that store pointers
with dev_set_drvdata - btt, pmem, etc.
On Wed, May 20, 2015 at 9:59 AM, Elliott, Robert (Server Storage)
<[email protected]> wrote:
>
>
>
> > -----Original Message-----
> > From: Linux-nvdimm [mailto:[email protected]] On Behalf
> > Of Dan Williams
> > Sent: Tuesday, April 28, 2015 1:25 PM
> > To: [email protected]
> > Cc: Neil Brown; Greg KH; [email protected]
> > Subject: [Linux-nvdimm] [PATCH v2 07/20] libnd, nd_dimm: dimm driver and
> > base libnd device-driver infrastructure
> >
> ...
> > diff --git a/drivers/block/nd/dimm.c b/drivers/block/nd/dimm.c
> > new file mode 100644
> > index 000000000000..a4c8e3ffe97c
> > --- /dev/null
> > +++ b/drivers/block/nd/dimm.c
> > @@ -0,0 +1,93 @@
> > +/*
> > + * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
> > + *
> > + * This program is free software; you can redistribute it and/or modify
> > + * it under the terms of version 2 of the GNU General Public License as
> > + * published by the Free Software Foundation.
> > + *
> > + * This program is distributed in the hope that it will be useful, but
> > + * WITHOUT ANY WARRANTY; without even the implied warranty of
> > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> > + * General Public License for more details.
> > + */
> > +#include <linux/vmalloc.h>
> > +#include <linux/module.h>
> > +#include <linux/device.h>
> > +#include <linux/sizes.h>
> > +#include <linux/ndctl.h>
> > +#include <linux/slab.h>
> > +#include <linux/mm.h>
> > +#include <linux/nd.h>
> > +#include "nd.h"
> > +
> > +static void free_data(struct nd_dimm_drvdata *ndd)
> > +{
> > + if (!ndd)
> > + return;
> > +
> > + if (ndd->data && is_vmalloc_addr(ndd->data))
> > + vfree(ndd->data);
> > + else
> > + kfree(ndd->data);
> > + kfree(ndd);
> > +}
> > +
> > +static int nd_dimm_probe(struct device *dev)
> > +{
> > + struct nd_dimm_drvdata *ndd;
> > + int rc;
> > +
> > + ndd = kzalloc(sizeof(*ndd), GFP_KERNEL);
> > + if (!ndd)
> > + return -ENOMEM;
> > +
> > + dev_set_drvdata(dev, ndd);
> > + ndd->dev = dev;
> > +
> > + rc = nd_dimm_init_nsarea(ndd);
> > + if (rc)
> > + goto err;
> > +
> > + rc = nd_dimm_init_config_data(ndd);
> > + if (rc)
> > + goto err;
> > +
> > + dev_dbg(dev, "config data size: %d\n", ndd->nsarea.config_size);
> > +
> > + return 0;
> > +
> > + err:
> > + free_data(ndd);
> > + return rc;
> > +
> > +}
> > +
> > +static int nd_dimm_remove(struct device *dev)
> > +{
> > + struct nd_dimm_drvdata *ndd = dev_get_drvdata(dev);
> > +
> > + free_data(ndd);
> > +
> > + return 0;
> > +}
>
> It would reduce the slight window for stale pointer usage if you add:
> dev_set_drvdata(dev, NULL);
>
> before
> free_data(ndd);
>
> in both nd_dimm_remove and the err exit for nd_dimm_probe.
>
> The same comment applies to all the drivers that store pointers
> with dev_set_drvdata - btt, pmem, etc.
>
That is automatically handled by disabling a driver. If some other
code may race with this it needs to hold device_lock(), which we do in
places.
> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On
> Behalf Of Dan Williams
> Sent: Saturday, May 16, 2015 10:22 PM
> To: Elliott, Robert (Server Storage)
> Cc: [email protected]; Neil Brown; Greg KH; Dave Chinner; linux-
> [email protected]; Andy Lutomirski; Jens Axboe; H. Peter Anvin;
> Christoph Hellwig; Ingo Molnar
> Subject: Re: [Linux-nvdimm] [PATCH v2 19/20] nd_btt: atomic sector
> updates
>
> On Sat, May 16, 2015 at 6:19 PM, Elliott, Robert (Server Storage)
> <[email protected]> wrote:
> >
...
> 2 items to check:
>
> 1/ make sure you have a your btt sector size set to 4k which cuts down
> the overhead by a factor of 8.
>
> 2/ boot with nr_cpus=256 or lower.
>
> Ross noticed that CONFIG_NR_CPUS is set quite high on distro kernels
> which revealed that we should have been using nr_cpu_ids and percpu
> variables for nd_region_acquire_lane() from the outset. This fix is
> coming in v3.
My system does have CONFIG_NR_CPUS=8192. I'll try the items you
suggested.
????{.n?+???????+%?????ݶ??w??{.n?+????{??G?????{ay? ʇڙ?,j??f???h?????????z_??(?階?ݢj"???m??????G????????????&???~???iO???z??v?^?m???? ????????I?